A non-statistician's view of the E-M exit poll controversy

[Time for change]

There has been a great deal of highly technical discussion and argument on the DU and elsewhere over the proper interpretation of the Edison-Mitofsky report http://exit-poll.net/election-night/EvaluationJan192005.pdf ,which purports to explain the wide discrepancy between the E-M polling results and the official 2004 Presidential election results. This discussion and argument is needed. However, one problem with it is that much of the discussion is too technical for non-statisticians to fully understand. I believe that it is very important for non-statisticians to understand the essential issues of this controversy, because the great majority of U.S. citizens are non-statisticians, and they need to understand this issue. This post is my attempt to frame the discussion in a manner that all DUers will be able to understand (although I welcome criticism by statisticians).

Although I am not a statistician I have taken several statistics courses, and I work with statistics (though not as complex as some of the discussion recently appearing in this forum) almost every day in my work as an epidemiologist for the FDA. I have read the E-M report and the USCV responses to it, and I strongly believe that the good majority of DUers are capable of understanding the crucial issues involved if they are discussed in lay language.

In the interest of full disclosure, I admit that I am a partisan Democrat, and that I believe that if the United States remains under the control of a Republican President, Congress, and judiciary for an extended length of time the consequences for the great majority of people in this country and the world will be terrible. However, I don’t believe that that makes me biased in my interpretation of the exit poll evidence or any other evidence bearing upon the possibility of 2004 election fraud. I say this because I recognize that in order to correct a problem (in this case, a Republican U.S. President and Congress) one must first understand the reason for the problem. So if I and thousands (or millions?) of others are wrong in believing that John Kerry won the 2004 election, and that George W. Bush was installed as President because of election fraud, then our efforts to expose this fraud may not only be wasted but actually be diverting us from our main goal, which is to bring a better government back to our country. On the other hand, if we are correct in our view, then fixing our election system is our best (and very likely only) hope of winning our country back, and I believe that exposing the election fraud is a very important step towards that goal. Therefore, it is in the interest of almost all of us, not to prove a specific point of view, but to investigate this matter until it is fully understood, and presented to the American public for all to see.

Having said that, I will also they that I am not 100% convinced that John Kerry really won the 2004 election. Maybe 97% -- so I continue to keep an open mind on the subject. This is based both on the exit poll evidence, which I discuss below, and also on a lot of other evidence, which I will not discuss in this post.


Here is my view and explanation of the exit poll controversy:

The 2004 E-M national exits polls predicted very different results than the official Presidential election results. Whereas Bush won the official results by 2.5%, the exit polls predicted a Kerry victory nationally by 3% -- a 5.5% difference. In addition, state exit polls predicted a Kerry victory in four states that Bush won – Ohio, Iowa, New Mexico, and Nevada – and a virtually even race in Florida, which Kerry lost officially by 5%. Of these states, the difference between the exit polls and the official results were statistically significant only in Ohio and Florida. In Ohio, Kerry lost officially by 2.5%, while winning the exit poll by 4.2% -- a difference of 6.7%. Winning either Ohio or Florida would have meant an electoral victory for Kerry. On the other hand, none of the states that Kerry carried in 2004 were predicted in the exit polls for Bush. None of this is controversial or denied by Mitofsky.

When exit polls differ substantially from official election results, there can be only three reasons (or combination thereof):
1. Random error, or chance
2. Biased polls
3. Impaired election integrity
This is not a controversial statement.

Given all that, here are the reasons, taken together, why I believe that the Mitofsky report supports the likelihood of election fraud, notwithstanding his statement in the executive summary of his report that “Exit polls do not support the allegations of fraud due to rigging of voting equipment”.


1. The role of random error (or chance)
The first step in the assessment of any statistical discrepancy is to assess the role of chance in producing the discrepancy. The likelihood of the discrepancy between the national exit polls and the official national results occurring by chance has been estimated by Jonathan Simon and Ron Baiman http://www.freepress.org/images/departments/PopularVotePaper181_1.pdf as being close to one in a million. The original response to the E-M report by US Count Votes (USCV) http://www.uscountvotes.org/ucvAnalysis/US/USCountVotes_Re_Mitofsky-Edison.pdf estimated that the likelihood of the discrepancy between the combined state exit polls and the official state results occurring by chance was about one in ten million. A proper combined likelihood of such a large discrepancy in both the national and state polls would multiply those two numbers, to give a result of one in ten trillion. Although the exact number can be and has been computed in different ways by different investigators, nobody, including Mitofsky, disputes the fact that the likelihood of this discrepancy occurring by chance is so small that it should not even be considered.

So that leaves two possibilities: Exit poll bias or impaired election integrity.


2. An initial look at the potential role of exit poll bias
Exit poll bias can be broken down into two components: Biased sampling of precincts AND bias within precincts – referred to as “within precinct error” (WPE). The former can be easily tested, and the latter cannot be easily tested (and many question whether or not it can be accurately tested at all). Mitofsky tested bias due to sampling of precincts and concluded that this bias actually favored Bush. Therefore, the hypothesis that the exit poll bias (which would have to favor Kerry) might explain the discrepancy between the exit polls and the official election results becomes less likely, since all of this bias must be concentrated within precincts (WPE), and this bias must account for not only the discrepancy between the poll results and the official election results, but also the bias due to sampling of precincts, which works in Bush’s favor.


3. Mitofsky’s explanation for the bias – the reluctant Bush responder (rBr) hypothesis
Mitofsky asserts in the executive summary of his report, as an explanation for the bias (henceforth referred to as WPE) that “Kerry voters were more likely to participate in the exit polls than Bush voters”. This has been referred to generally as the reluctant Bush voter hypothesis, otherwise known as rBr. However, Mitofsky does not refer to this as a hypothesis – he simply states it as a fact. Not only that, but neither in the executive summary nor in the body of the report does he provide ANY evidence to support that contention.


4. Testing the rBr hypothesis by looking at response rate by partisanship
If the rBv hypothesis was valid, where would you expect the lowest voter response rates to be? I would think that the lowest voter response rates would be most likely to occur in precincts that leaned heavily to Bush – since the hypothesis postulates that the reason for the biased polls is reluctance of Bush voters to participate in these polls, more Bush voters should mean lower overall response rate.

However, when USCV analyzed the data presented in the Mitofsky report, they found exactly the opposite: Precincts with the highest percentage of Bush voters had the highest, not the lowest response rate. This must certainly count as evidence against the rBv hypothesis.


5. Testing the revised rBr hypothesis by looking at WPE by partisanship
But don’t yet give up hope on the rBr hypothesis. It can be revised to say that, although Bush voters in general were more reluctant to participate in the polls than Kerry voters, this did not apply to precincts where there were a very high percentage of Bush voters, because in those precincts the Bush voters would perhaps feel more comfortable participating in a poll.

This revised hypothesis can also be tested. If the rBr hypothesis applied only to precincts without a heavy preponderance of Bush voters, then one would expect that those precincts would be where the bias (i.e., WPE) would be found. But in fact, by Mitofsky’s own data, precisely the opposite is the case: The average WPE is highest, not lowest, in precincts where there were a very high percent of Bush voters (80% or more). This too must count as further evidence against the rBr hypothesis.


6. Consideration of election fraud by assessment of voting machine type
As noted above, Mitofski dismisses the possibility of election fraud in his executive summary by saying that his polling found no evidence for it. To support this contention he notes that they “… found no systematic differences for precincts using touch screen and optical scan voting equipment”. This is the sum of his evidence for the absence of election fraud.

But both touch screen and optical scan machines count the votes by computer. And so do all other methods of vote tabulation except for the hand counting of paper ballots. Deep in the report, but not in the executive summary, is the average WPE data by type of voting equipment:
Paper ballot: -2.2
Mechanical voting machine: -10.6
Touch screen -7.1
Punch cards -6.6
Optical scan -6.1

Note that the average discrepancy (as manifested by a more negativeWPE) is considerably less in precincts where paper ballots were used, compared to any other method. USCV reported this finding but did not make a big deal of it, because they didn’t feel that it was conclusive, and they were concerned about the possibility of over-interpretation.

There are two problems with attributing this finding to election fraud. First, there were no tests of statistical significance presented by Mitofsky. It is not possible to assess the statistical significance (i.e., role of chance) of this finding without the raw data, and Mitofsky is not releasing the raw data to anyone. Secondly, it is possible that this data might be “confounded” by some other factor – for example, the WPE was less negative in rural areas than in urban areas, and rural areas were more likely to use paper ballots, and therefore, the use of paper ballots in rural areas could partially explain this finding. I find this somewhat unlikely, partly because the spread in WPE between urban and rural areas was less than the spread between paper ballots and other types of voting. Therefore, the use of paper ballots in rural areas could not possibly explain all of this discrepancy by voting machine type. However, it could explain some of it. Again, it is impossible to test this hypothesis without access to the raw data. Nevertheless I certainly believe this finding to be suggestive of election fraud.


7. Comparison of election fraud by assessment of swing states versus non-swing states
If the 2004 Presidential election was fraudulent, one would expect more fraud to have occurred in those states where there was a reasonable chance of switching their electoral votes from Kerry to Bush (i.e., the swing states). Of the 11 main swing states (OH, FL, PA, WI, MN, NM, IA, NV, NH, MI, CO), according to Mitofsky, in five of them there was a discrepancy between the exit polls and the official election results that were outside of the margin of error (I’m defining outside of the margin of error as less than a 5% probability of occurring by chance). These five swing states included Ohio, Florida, Pennsylvania, Minnesota, and New Hampshire. Of the remaining 39 states, only 8 were outside of the margin of error (In all 13 states that were outside of the margin of error, the exit polls favored Kerry, compared to the official election results). I believe that this finding also supports the suggestion of election fraud.


8. Comparison with previous elections
Mitofsky admits in his executive summary that the “exit poll error” was higher in 2004 than in previous years for which he has data (going back to 1988). I have a hard time believing that this higher discrepancy in the 2004 election and its concurrence in time with a greater ability than ever to use secret software codes to fix elections is just a coincidence. It’s possible, of course. But when taken together with everything else, I think it is very suspicious.


9. The controversy between USCV and Febble
I mention this last issue only because there has been so much discussion on this, and I would like to attempt to put it in perspective. I seriously do not believe that this issue has much importance, compared to any of the issues noted above, and therefore I think that spending too much time on this may be harmful to our efforts to shed light on the main issues.

The controversy over this issue has involved very complex statistical concepts. Because of the complexity of the issue, the language used, and the sheer volume of material, I have been unable to find the time to fully understand this argument, and even if I did have the time I’m not sure that I could understand it well enough to have a valid opinion on who’s right and who’s wrong.

But I do believe that I understand it well enough to know that it has peripheral importance to the central issues, as discussed above. USCV (or at least some of their members) has claimed that the data support going beyond what I’ve discussed above. Specifically, they have devised complex simulations to show that the rBv hypothesis is untenable. Febble, on the other hand, claims that the rBv hypothesis is possible, based on her computations (I don’t believe that she goes beyond “possible”, but I don’t fully understand exactly what she is trying to say, so I may be wrong about that.) In any event, whether the rBv hypothesis is untenable or merely possible, certainly the above noted considerations cast considerable doubt on it.


Summary

For all of the above reasons, plus the existence of volumes of non-exit poll related evidence, I strongly believe that John Kerry won the 2004 Presidential election. But even if others may find the evidence weaker than I do, certainly there is enough evidence to cast enough suspicion on the integrity of the election to warrant serious investigations into this all important issue, as well as concurrent vigorous efforts to improve the integrity of our election system. This effort should include, among many other things, public access to the raw exit poll data, as well as the secretly programmed voting machines that determined the results of our last Presidential election.

[OnTheOtherHand]

1. outstanding post

At some point I owe you an explanation of why I'm not 97% convinced that John Kerry won the election. (And I hope we'll continue to sort out the issues about releasing exit poll data.) But this is a great summary of many of the important issues. And yeah, we have so much work to do.

[Time for change]

2. Well, thank you very much

You don't "owe" me an explanation for your doubts about John Kerry winning the election. But I'd very much like to hear it anyhow.

[Birthmark]

3. Amazing post.

I'm sure that it will be helpful to those of us who aren't satisticians. I will probably re-read this several times before I post a response to the substance.

[understandinglife]

4. "...enough evidence to cast enough suspicion on the integrity of the ....

.... election to warrant serious investigations into this all important issue....

And, I agree with the reasonable conclusion of your analysis "For all of the above reasons, plus the existence of volumes of non-exit poll related evidence, I strongly believe that John Kerry won the 2004 Presidential election."

Indeed.

Spectacular post. Thank you.


www.missionnotaccomplished.us - Tragically on the path to MNA Day 4

[spooky3]

5. Thank you for a great public service.

[Peace Patriot]

6. Many thanks, Time For Change! This summary is indeed outstanding.

I'll leave it for the statisticians to vet it for exactness. I am not a statistician, but I am reasonably intelligent and have read and re-read just about everything--going back to Jonathan Simon's initial reports and Steven Freeman's first paper. (I've even read large parts of the Febble-USCV discussion which, I agree, is a side issue--except on the political front, that is, what Edison-Mitofsky might be up to in its highly questionable handling of the data, and Febble's late disclosure that she is working for Mitofsky).

Your summary of all this material is not only excellent, it fills a critical need in helping non-experts to understand it and to grasp its importance. It is comprehensive and clear. Thank you so much!

I find your point #7 (swing vs nonswing states) the most compelling, and I always have. If the exit polls were biased, they should be biased in a more or less evenly distributed way, given the size of the sample. That the "bias" occurs just where Bush needs it is extremely suspicious, and until THAT is explained, there is really no other conclusion one can come to, except election fraud.

Mitofsky's 'rBr' only addresses a possible political bias. It does not address a "swing state" bias. He hasn't even tried to address that, as far as I know.

My thinking goes like this. The skew to Bush in the official tally vs. the exit polls is basically impossible (ten million to one odds against it). AND Kerry won the exit polls.

(If it was just a Kerry win of the exit polls, one could certainly imagine a bias that could produce that result. But there is no conceivable bias that could produce the swing state skew. And I wish election fraud writers would emphasize this more. To me it seems like a slam dunk.)

------

There are three additional factors in the election that, when combined with the above, drive me to a 99% conviction that the election was stolen (always leaving a little room for doubt--in all things).

1. The fraudulent election SYSTEM. This election system didn't occur by accident. It was deliberate Republican policy NOT to have a paper trail (Tom Delay blocked it in Congress), and to short the oversight group (the Election Assistance Commission) on funds. There were large amounts of money infused into the states, and heavy lobbying by electronic voting companies, pushing the states into precipitous purchase of unreliable, insufficiently tested, insecure, hackable election systems--in which "no-brainer" controls (such as open source code, and a paper ballot backup) were quite deliberately omitted. Also, the companies that counted all our votes in secret are run by major donors to Bush and to rightwing causes. The conditions for an honest, verifiable election were not present. The election system itself was a fraud--not just who "won."

2. VIRTUALLY ALL ELECTION "GLITCHES" FAVORED BUSH. In 86 out of 88 reported incidents, electronic touchscreens changed Kerry votes to Bush votes (huge odds against this--a "smoking gun" pointing at computer programming). Other "malfunctions" included machines defaulting to Bush, machines erasing the Kerry vote in a party vote ballot, machines counting backwards eliminating thousands of votes, and machines reporting more votes than voters--all favoring Bush and hurting Kerry. This is in addition to several statistical studies showing a strong tilt to Bush in non-paper ballot votes (in No. Carolina, Florida, Ohio and overall). Other "glitches" included shortages of precincts and of voting machines only in Democratic areas, misdirecting voters, intimidating voters, unfairly challenging voters, 'losing' thousands of absentee ballots in Democratic areas, and shredding Democratic voter registrations--again, all favoring Bush and hurting Kerry.

2. INTENT. See #1. (If the Bush regime had wanted a verifiable election, why didn't we have one?). Plus: a) Massive violations of the Voting Rights Act against minority and other Democratic voters, by Republican election officials in Ohio, Florida and other places, with NO Justice Department effort to prevent them or to investigate them when they happened. The Justice Department is under Bush regime control. These violations were widespread and egregious. Since when do we permit such things in the U.S. of A.? And, b) The Bush regime stole the previous election (in 2000), and there is also evidence that they stole at least one Senate seat (Georgia) in 2002. The installation of an unverifiable election system--when we could easily have had a verifiable one--the many overt acts of vote suppression in 2004, and previous history, point to an intent to defraud the voters of an honest election.

3. PREDICTABLE KERRY WIN. The Democrats had a blowout success in new voter registration (nearly 60/40) in 2004. Most new voters voted for Kerry. Most Nader voters voted for Kerry. Most independent voters voted for Kerry. Most Gore 2000 voters voted for Kerry (and had much motivation to do so, having been cheated in 2000, and this time getting all the non-voters among their family, friends and co-workers to register and vote--resulting in the aforementioned Democratic voter registration success). So where did Bush's margin of victory come from? (Karl Rove now says they had an "invisible" get out the vote campaign. Right.) The exit polls show a 3% Kerry win AFTER all the vote suppression had winnowed Kerry's vote (the Kerry voters who never got to vote--and thus were never exit polled). If you add in an estimate for the disenfranchised, his real margin was even higher than that (4% to 6% would be my guess).

And there is a fourth thing that really nails it for me. Bush's approval ratings. 42% today. 49% on the DAY of his 'Inauguration.' This is unprecedented disapproval for a recently "reelected" 2nd term president. In addition, opinion polls show huge disapproval of all major Bush policies, foreign and domestic, up in the 60% to 70% range.

You can't just look at numbers--although the numbers in this case are riveting. You have to look, too, at likelihood--at means, motive, and opportunity, at intention, and at a previous history of similar behavior. What would a regime with a history of at least one other stolen election, who had perpetrated an unjust and illegal war on a pack of lies--a war that killed over 100,000 innocent people--and who were stealing the country blind, do? Would they steal an election?

And, finally, in the realm of feeling and intuition, you have to make a judgment of the American people. 63% of them oppose torture UNDER ANY CIRCUMSTANCES. Nearly 60% of them oppose the Iraq war today, as they did prior to the invasion. Their jobs have been outsourced. Their schools, fire departments, libraries, medical care, supports for the poor and the downtrodden, hourly wages, and everything good about America are getting severely squeezed for funds and are deteriorating, while fat cats raid federal and state coffers to enrich themselves and to secure Bush Cartel control over Iraq's oil. Would these people have the common sense to vote the Bush Cartel out? Would they be able to see through all the propaganda--a people who, despite their 'terrorist' fears, expressed 63% disapproval of torture UNDER ANY CIRCUMSTANCES?

I think they would, and they did.

----

And then Edison-Mitofsky cheated them out of knowledge of their victory by FALSIFYING the exit polls (Kerry won) to fit the official result (Bush won) on everybody's TV screens on election night. The worst journalistic crime ever committed, done in cahoots with the news monopolies. Stifling protest and investigation. Denying the American public strong evidence of election fraud. And inflicting the Bush Cartel upon us once again--a regime that the majority of Americans rejected in their vote and continue to strongly reject in opinion polls. And they don't even know why they are not being listened to.

[OnTheOtherHand]

7. swing states

There's a lot in Peace Patriot's post I agree with, but I need to comment on the swing state point. (EDIT: also the last point.)

That the "bias" occurs just where Bush needs it is extremely suspicious, and until THAT is explained, there is really no other conclusion you can come to, except election fraud.

Two things.

First of all, I have never, ever seen a p value reported for the oft-cited result that 5 out of 11 battleground states were outside the margin of error, compared with only 8 out of 39 other states. (I'm not saying that no one ever did it!) I just calculated the chi-square, and the p to four digits is 0.0958 -- a bit under 10%. That isn't a slam dunk.

Second, Febble shared a scatterplot with me a while back, but we can redo it from TIA's numbers and see if it turns out radically differently. (There are somewhat different ways of measuring the gap between official vote totals and exit poll results, and I'm not sure what she used.) It shows a pretty substantial relationship between Kerry margin and pro-Kerry exit poll 'bias.' The places where Kerry's official totals were the farthest below his exit poll totals tended to be heavily Kerry anyway, and the places where Bush's official totals were the farthest below his exit poll totals tended to be heavily Bush. When you look at all 50 states together, the battleground states don't really seem to stand out. (New Hampshire somewhat does, but then, other northeast states have high 'bias.')

So, the bias occurred in lots of places, including places where Bush didn't need it and couldn't appreciably benefit from it (at least in the sense of picking up states -- and, in the case of Vermont, I don't suppose it had much impact on the popular vote, either).

Now, if Bush stole Ohio, I don't much care what happened in the other 49 states. But I think the overall swing-state argument is less compelling than you have made it out to be.

That said, do I have a nice story about why the bias seems to be more pro-Kerry in pro-Kerry states, etc., even though there doesn't seem to be a strong relationship at the precinct level? No, I do not.

(EDIT:)

Look, I don't work for Mitofsky, I don't know Mitofsky, we've traded maybe three e-mails in the last three years. The exit polls aren't designed to audit elections. Maybe we wish they were, but they're not. The history of election polls is one of progress being made through big mistakes ("Dewey Beats Truman!"). I firmly believe that on election night, Mitofsky wasn't thinking "Wow, Bush is stealing the election, how do I cover for him?" He was probably thinking things like "Whoa, this is a slow-motion nightmare suddenly happening very fast" and "Lucky we didn't actually call any states the wrong way" and "Gee, am I going to have some unpleasant conversations with network people." No way do I think Mitofsky set out to falsify the election results. You can draw your own conclusions (and you certainly don't have to agree with his conclusions about the election, or my conclusions about him), but I thought I should say.

[autorank]

12. A Reply to Your Doubting & Denying

You opine: "I firmly believe that on election night, Mitofsky wasn't thinking "Wow, Bush is stealing the election, how do I cover for him?" He was probably thinking things like "Whoa, this is a slow-motion nightmare suddenly happening very fast" With all due respect, that and $2.00 gets you a cup of coffee at Starbucks. The questions about Mitofsky and EM raised here are not based on gut feelings, they're based on a ton of evidence presented and discussed here. Here is one DU special. Mitofsky has been using the Reluctant Bush Responder excuse for years. Reluctant Bush Voters--A long time Mitofsky Excuse That is the type of information that's useful in evaluating how Mitofsky handles himself. He's clearly bending over backward to do anything he can to legitimize the Bush victory, including trashing himself Mitofsky trashes himself to legitimize Bush win. Mitofsky is disparaging his own reputation to deny the evidence from the national exit polls he conducted that clearly show election fraud and a Kerry victory.

You are a frequent proponent of Bush winning, which means you need to invalidate entirely the use of national and state exit polls. The rhetoric of denying fraud and a stolen election is a two step process. In step one on DU, Febble (a FOM, Friend of Mitofsky) argues the statistical case, the rBv, reluctant Bush voter, which has been totally discredited here rBv Falls. Step two of the denial process is the statement that exit polls, as you say, "are not designed to audit elections." This is a crafty rhetorical device. Of course they're "not designed" to do this, that's a truism. They're designed to poll the sentiment of the voting public. This in no way precludes them from being used to spot fraud in elections, exactly what has been done so well in this post by TimeForChange and all over DU Elections by TruthIsAll.

While you seem to deny the utility of exit polls, you challenge the Blue-Red Shifts which violate statistical analysis. This is of interest since you seem to support the Bush win with the exit polls that you try to dismiss. Nevertheless, with regard to the blue-red shift, there is ample documentation for the unlikely nature of the changes. Look here More on States Bush Took & Kerry Won and here "Red Shift--Z scores, Another Way to Look at It

I hope it's time for you annual Mitofsky email exchange. Why don't you contact him and tell him: "Release the NEP raw data to the public with a verification process to establish if any changes have been made subsequent to the elections." Now that would be useful for those of us who would like to stop election fraud and establish with finality the fact that Bush stole the election.

You might also tell him he's now the top eFd (enthusiastic Fraud denier) on DU

[autorank]

17. Help, I'm stuck in a Mitofsky sandwitch!

[OnTheOtherHand]

19. doubting vs. denying

autorank, I won't spend time trying to convince you of anything, and I won't be able to chase you all over DU responding each time you write something about me that is, umm, not quite right. But a comment on this "enthusiastic Fraud denier" business.

Doubting is my job, and it's different than denying. IMHO doubters and believers can work well together; doubters alone, or believers alone, don't get much done, or do the wrong things.

That Mitofsky has "been using the RBR excuse for years" could prove that Bush owns him -- although if Bush did own Mitofsky, I still think Mitofsky could have found a way to rig the exit poll up front (should be even easier than rigging the election!!) and spared himself and Bush a lot of trouble. It could also mean that he assumes that when his surveys clash with official results, probably it's his surveys that are wrong. That is the assumption that most survey researchers go by. It doesn't mean he is right.

I don't think my comments on swing states "support the Bush win with the exit polls that I try to dismiss." They are a direct response to a comment up-thread. I will check out those links when I can. I never asserted that the swing-state exit poll results support the Bush win, per se.

Based on what little I know about data security, I think it's way too late to establish an audit trail for the exit poll results. Recounting op-scan ballots seems more workable. More elsewhere on what Mitofsky should release.

[autorank]

20. Doubting followed by affirmation--a real opportunity.

OTOH (you now have an acronym after only a few messages),

You've convinced me you're a doubter (don't know if it's hopeless). I wasn't aware of commenting on OTOH elsewhere on DU. I have commented on Mitofsky's results and behavior, however, so feel free to respond to those when you run into them.

You said, "IMHO doubters and believers can work well together; doubters alone, or believers alone, don't get much done, or do the wrong things." Well I agree but only up to a point. IMHO, the case has been made for extreme doubt regarding the legitimacy of the 2004 results and it's time to move forward to save the country. Recursive doubt is not useful as you engage in this type of fight, I'm sure you will agree.

I'll admit to being a polemicist, at times, but let me suggest a few points that really bother me.

--First the Mitofsky data should be released for public review just as soon as possible in a form that's verifiable. This would settle so many questions and provide a basis on which to judge what happened in the election. There have been excuses offered elsewhere by others claiming it's proprietary, etc. The value of this data is so great because the well being of our country is at stake...release the data now. If released, the data provides an opportunity for great services. Mitofsky is no longer the "goat" in our small but growing doubter community, he achieves the status of an intellectually honest man and a patriot...an amazing combination.

--Second, much of my doubt centers around the time line of the exit poll data presented here by TIA. To me, it seems like the NEP went from being a poll trying to answer a question, to an answer used to retrofit a poll. It's my understanding that if you have the answer, a discriminant function may be your analytic tool. The answer was Bush, BUT everything up to the final poll showed Kerry. Why the change? How did the data suddenly transform to fit the results already known, Bush wins?

--Finally, I actually admire Mitofsky. I don't know him (although I may email him just to say I did..."I don't know Mitofsky but I have a one sided email relationship with him"). Why do I admire him? Because he pulled off this poll and got it right, because he's obviously skilled at mounting a complex operation and selling his services. What I find troubling is the "denial events," like the press release where he trashes himself and like the contrivances of rBv/rBr's...whatever. It is beyond my imagination (which is saying a lot) to see how anyone of Mitofsky's intellect and ability can willingly and repeatedly disparage his results. The contortions are painful to watch...my election night poll was right, the last one; but the the others showing Kerry winning are wrong due to my incompetence...Seriously, isn't this what he's saying? It's painful but I suspect he was and is under a lot of pressure. Imagine being the one to say, it was fixed, this guy stole it again.

So what do you say, will you email him and make a strong case for him to release the data in a verifiable form?

Despite being an extreme doubter on the last two elections, I'm actually an optimist on the outcome of the entire process of reclaiming our democracy and national sanity. I can see a situation where Mitofsky comes forward and explains everything, including having it absolutely correct through most of election night and the events that led to the mystery reversal. This is one case where doubting only delays the inevitable. When you lead, at least you have some control of the events.

[OnTheOtherHand]

23. thanks for a really interesting post

(Only problem is, I've dropped my work to answer it!)

I absolutely agree that recursive doubt is self-swallowing. I'm probably coming across as more extremist than I am because -- well, never mind. Let's just say May has been a very strange month.

Generally, I would say, we need to figure out every way that the election could have been stolen, and try to slam those doors. I will criticize 2004 fraud arguments if I think they aren't compelling, but for heaven's sake, I'm not telling anyone to relax. We have so much work to do.

(1) I think Time for Change agrees with me that there is a legitimate privacy concern about just flat-out releasing everything. But I swear to God, I am not an apologist for just flat-out releasing nothing. One particular project I have for the next month, depending on how fast stuff comes out of ESI, is to figure out whether the "blurring" thing they did in Ohio is legitimately useful. If it is, then I will be raising funds to make sure that it happens for other states as well. I will also be among the pack of people leaning on Mitofsky to release more analytical results.

(I'm not sure what you mean by "release the data in a verifiable form." So I don't know whether "blurring" will count for you, or not. As I think I said upthread, I don't think there is any conceivable way for E/M to prove at this point that there hasn't been weird tampering with the survey -- unless they built an audit trail into the system, which I doubt. Bar codes on each survey? Tracking the blank ones?)

So I don't know whether I've said yes to exactly what you asked, but I will put it this way: I will work in every way I can think of to get every bit of information released that doesn't raise real problems for privacy or future exit polls.

(2) You do seem to think (am I right?) that Mitofsky is sure that the exit poll got it right the first time -- that Kerry won. But I don't think that, because that's not the way survey researchers are trained to think. When you "get the election wrong," your first impulse isn't to blame the election, and in fact you look pretty pathetic if it is. Now, there are lots of examples in the history of science of people discounting their data when they should have discounted their theories instead, and Mitofsky could be in that boat. No matter what the right story is, it sure is interesting to imagine being Mitofsky.

(3) If you'll grant me for a moment, as a thought experiment, that Mitofsky didn't think on election night that the exit polls proved Kerry had won, then what part of the timeline troubles you the most? If you stuck a thermometer in your mouth and it said 94, would you assume that you were hypothermic, or would you add four degrees? (Admittedly it's a bit different if you hand-crafted the thermometer!)

I'm a great optimist, too. We'll need it, since it will take a while to turn the ship around. Hang in there.

[spooky3]

24. The raw data should be released to responsible researchers.

Social scientists who publish in peer-reviewed journals are expected to keep their raw data for a period of years (the period varies by field) in order to make them available to those who would question the findings. They can prevent others from abusing this sharing, e.g., publishing other articles from the data that have nothing to do with questioning the original findings, so this is not a good reason to refuse access.

I realize that Mitofsky has not presented his data in a peer-reviewed journal but it seems to me that the principle is the same, and that given the important public policy issues at stake, there is an even stronger argument to be made that he is ethically obligated to make the raw data available to people who will handle the data appropriately.

[autorank]

30. Doing & Thinking...thinking outside of the box.

Thanks for the response. In the spirit of avoiding work and addressing the larger issues at hand (i.e., our future), I have a few comments.

Apparently we agree more than we disagree.

May has been a very strange month, hasn't it? Things are just starting to creep over the horizon, the fantasies are being burned away by the light of reason leading to truth.

You said: "I'm not sure what you mean by "release the data in a verifiable form."
--I'm talking about the type of examination that's done when a bank is audited or when internal auditors from headquarters show up at a subsidiary to do a full financial audit. We need to know that the data is what it was and know how it's been preserved over time (like a 'chain of evidence') in order for any independent analysis to be meaningful. Another example, when accreditation commissions show up at hospitals, they have the the right to review everything including patient charts. They are bonded in a way that compels privacy and so forth. This is highly sensitive data also, maybe as sensitive as who got polled, and it's a procedure that works well. I'm sure there are ways to do in depth due diligence in a way that assures that the data released and analyzed has the integrity required for the conclusions that will follow.

You said: "Now, there are lots of examples in the history of science of people discounting their data when they should have discounted their theories instead, and Mitofsky could be in that boat."
--Well, I'm now looking for my tattered, aged copy of Kuhn's "The Structure of Scientific Revolution" which makes this same point. Mitofsky's evening did not occur in a vacuum. He knew about the various events in Ohio and elsewhere to game the system. He also knew about the pre-election polls (Zogby et al) and the Bush approval rating and the role that's played in presidential outcomes. If ever there was a time to think outside the box, discard old theories because of anomalous evidence, that was the time. It's a shame TIA wasn't there to help him. (I'll grant you that Mitofsky may not have known that the Redskins loss to the Packers was an unchallenged past predictor that showed Kerry would win. I was there...rooting for the Packers.)

You said: "No matter what the right story is, it sure is interesting to imagine being Mitofsky."
--I wonder if that's like "Being John Malcovich?" At your prompting I did imagine that. If I were him, I'd get that data in the public domain quickly and in a process over flowing with integrity. That would shift the focus (and heat) from Mitofsky to an horizontally arrayed set of analysts and take the heat off.

You said: "If you'll grant me for a moment, as a thought experiment, that Mitofsky didn't think on election night that the exit polls proved Kerry had won, then what part of the time line troubles you the most?"
--All of it...especially the retrofit at after 12:25 when the numbers went from Kerry wins 51-48 to Bush wins. Very troubling. The alarms should have gone off.

You asked: "If you stuck a thermometer in your mouth and it said 94, would you assume that you were hypothermic, or would you add four degrees?"
--I'd be in my car headed to the ER...promptly. Even if I'd "hand crafted" the thermometer, larger laws would come to bear like physics, chemistry, biology. Mitofsky 'hand crafted' the exit poll I presume but he did not invent the rules governing the statistical analysis of evidence. Fun experiment, though.

Go for it, turn the ship around...I support that 100%

[Time for change]

34. This argument about Mitofsky's motives is very interesting

I know almost nothing about him, except what I know from reading his report. It's obvious from reading my original post that I don't think much of his report, and obviously that reflects on him.

But still, I don't know what kind of pressures he was under, and I usually don't like to attribute base motives to people until I understand the situation better. And as OTOH says, it could have been worse -- if he was totally crooked he might have cooked up his exit polls to begin with. Well, actually I don't know if he could have gotten away with that or not.

But what I really wanted to say is that I have my own theory about Mitofsky, which I have never heard anyone mention:

Why would a leading international expert write such a god-awful report? More specifically, why would he reach various critical conclusions without sufficient explanation as to how those conclusions were reached, and why would he present numerous comparisons in his report (presumably for the purpose of supporting his conclusions) without even presenting statistical analyses to help in the interpretation of those comparisons? The only thing that I can think of is that he must have been under intense pressure not to give any hint that fraud might have been involved. Where that pressure came from, I don't know.

Here's the main point IMO: In the executive summary he states unequivically that his data provide no evidence of fraud. And the only fact he gives to support that contention is that the "bias" in precincts that used touch screens was similar to the "bias" in those precincts that used optical scan machines.

But then, buried deep in the report, he actually presents a table that shows that precincts that used paper ballots had a much less negative WPE than those that used any other method to count votes. To me, that is absolutely glaring. So I think that he was under intense pressure to announce in the executive summary that there was no evidence of fraud. But deep down in his heart, we wants the world to know, and that is why he presented the table later on in the report. That's the only thing that I can think of because certainly he couldn't have thought that the whole world would miss that. And he certainly didn't have to present that table.

[autorank]

39. I think you're on to something & a belated Welcome to DU!!

I love this line in your post, "But then, buried deep in the report, he actually presents a table that shows that precincts that used paper ballots had a much less negative WPE than those that used any other method to count votes. To me, that is absolutely glaring." Glaring indeed. This is like a Kabuki Theater those movies (e.g., Memento) with all the hidden (or not so hidden) clues. Your analysis makes perfect sense to me. We can only speculate about the intensity of the presumed pressure; it must have been chilling. If this is right, and I'll just assume it is, Mitofsky is like some 21st century existential hero maintaining his honesty in the only way he can; hoping for a discovery that will affirm that honesty. We are through the looking glass...

[Time for change]

64. Yes, we are indeed through the looking glass

Have you read "Into the Buzzsaw -- leading journalists expose the myth of a free press"? It's absolutely shocking, but all too true. I would highly recommend it for anyone on the DU, but even more for anyone else.

[helderheid]

44. yes

I think you may have hit the nail on the head.

[tommcintyre]

69. Great report! But Mitofsky is CLEARLY UNETHICAL AND DISHONEST

You did a great service taking the time to write all this up. Thanks!

And you revealed some interesting things about Mitofsky:

1) He claimed rBr as fact in his original report. At MOST, is was a hypothesis, he knew that, EVEN his rBr "supporters" call it that. Plain and simple, legitimate professionals NEVER claim a hypothesis as a fact - by definition, it is simply a "guess", or "hunch" that must be investigated further. THIS ACTION ALONE IS A STRONG BREACH OF ETHICS. Thanks for pointing it out.

2) As you also pointed out, he also advanced rBr WITHOUT ANY SUPPORTING EVIDENCE. Again, this is something no ethical professional would dream of doing, in a formal report, as he did. They would wait until the investigation was reasonably done, and ONLY THEN publish their findings. Again, Thanks for pointing it out.

In your post above, you say:
"I know almost nothing about him..."
When you find the time, it would be well worth your while to learn more about him. Here's some info to get you started:

---------------------------
3. I've listened to Victoria Collier's interview, and highly recommend it

http://www.edwardsdavid.com/BushVideos/infidelguy.com_Vote_Scam_a.mp3

Also, read the first eight chapters (three more than at the votescam site) of the book here:
http://www.constitution.org/vote/

*<Verification here:>
http://www.votefraud.org/Archive/Write/newhampad.htm
"...early media reports stated that Bush and Buchanan were running neck and neck in the 1992 New Hampshire primary..."

<My original source:>
http://www.votefraud.org/News/2000/7/071800.html
-----------------------------------------------

And this is just the tip of the iceberg. Mitofsky has been at this since the mid-sixties.

I've found a clear pattern of <from another post of mine:>

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x371175#372063
"Mitofsky uses a number of methods to "cover-up". It seems to range from computer meltdowns, "found votes", withholding the data, "convenient guesses", etc.

So what may have happened in 2004? <To "trip up" Mitofsky.>

- More scrutiny because of publicity of the 2000 theft.

- Internet leaks"

The best response Mitofsky's chief apologist (around here) could come up with was the: "But you can't prove it" ploy.
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x371175#372245

We've seen this over and over again. Whenever they are backed in a corner, they use this one in desperation.


I hope all this helps. Keep up the good work!

EDIT:
WHY might he have "presented the table later on in the report..."?

- sloppiness (he's gotten away with this for so long...)
- arrogance (see sloppiness above ;) )
- CYA (cover yer ass - just in case the efforts now in full swing are unsuccessful to quash the USCV report, he can say...)

I don't like to speculate on specific motives like this. I prefer to focus on the specific actions - and the obvious results. So, in the end, I must say I have insufficient data at this time to to say why he presented the table. But WHAT he is up to (and has been in the past) is pretty darn clear.

[Time for change]

74. Thank you for all the info Tom

You may be right about Mitofsky, I just don't know. Obviously you have been looking into this issue much longer than I have, and you know much more about it than I do.

I don't want to put words into anyone's mouth, but I think it's fair to say that no one who has posted on this thread thinks much of Mitofsky's report. But the extent of corruption implied in the material in your post is another matter. Of course at this time, with all that has gone on in this country since the 2000 election, my level of suspicion is high enough to suspect it as a real possibility.

VoteScam sounds like a very interesting and informative book that I should read, though I'd rather read it reclining in my chair than at the computer. Do you if it still selling at book stores?

[Time for change]

33. Much appreciate your efforts to get the needed data released

I don't know how you intend to do that, because I wouldn't know where to start. But it's great to know that you'll be working on it.

And yes, I do agree with you on the basic issue of the release of exit poll data, as exemplified in these quotes of yours and mine from another post:

You: "We might be able to get E-M to release enough data to conduct a "fairly" transparent analysis without compromising anyone's privacy. I'm just saying that if they release everything, I am pretty sure that some individual votes will be revealed."

Me: "I agree with that. They don't need to release everything, but certainly every attempt should be made to release enough data to conduct a fairly transparent analsysis without comprosing anyone's privacy."

[OnTheOtherHand]

42. no mojo up my sleeve

but if / to the extent that "blurring" works, then it becomes largely a fundraising issue. And to the extent that Mitofsky can release more analytical results, I can feed the grumbling among public opinion analysts (although they will be inclined to give him some time to operate -- by their standards, the election was not all that long ago).

Bear with me, I won't have anything deep to say about this for a while (well, I hope to have an opinion about "blurring" within a week or maybe two).

I'm on the road today (well, have to run to catch a bus!) so can't cover all the threads, but I don't disagree with you about the swing-state scatterplot -- although I do think it puts in perspective that the errors weren't biggest in the swing states, as some assume. (When stats people start arguing about one vs. two tails, it usually isn't a good sign! I don't consider p < 0.05 sacred. It seems like a pretty crude test regardless.)

[passy]

45. What a wonderful quote.

"I still think Mitofsky could have found a way to rig the exit poll up front (should be even easier than rigging the election!!) and spared himself and Bush a lot of trouble"
What if he had, but not done it well enough and had to fix it up one more time on Nov 3 (the 43% * voters from 2000 for example).
If we also imagine that the * mafia was going to steal the election they would base their fix on the result from the last election, but they somehow didn't anticipate such as huge Kerry win and thus they had to do some more work on Election night.
If we actually envisage that a double fix was applied in both cases we could be looking at a huge Kerry victory 10% or more.
Now don't say where is the proof because there isn't any right now,
but if we consider that over 90% of Gore voters, that 65% of independents and 75% of Nader voters voted for Kerry, what kind of margin are we looking at now?

[autorank]

54. You have a wonderful quote, sums it up

This is the verbal equivalent of the points TIA makes:

"over 90% of Gore voters, that 65% of independents and 75% of Nader voters voted for Kerry, what kind of margin are we looking at now?"

How stupid do these people think we are? (Don't answer that!).

Nice post.

[OnTheOtherHand]

60. "What if he had, but not done it well enough"

I can buy that sort of argument on the election fraud end, maybe, but I have a really hard time buying it on the exit poll rigging end. I figure if someone in the E/M core wants the national exit result to come out 51/49 Bush at any given moment, it comes out that way, period. It's not that hard. You don't have to choose your fudge factor in advance, you can choose your target and automatically figure the fudge factor. So, I figure E/M didn't decide in advance how they wanted the exit results to come out.

(Also, if Kerry actually won by 10 points or more, I'm not sure what was going on with all the pre-election polls.)

As for the rest, I don't know. (How do we know what % of Gore voters etc. voted for Kerry? exit poll? some other poll?)

[glitch]

63. This is my most likely scenario too - count riggers would rig polls,

of course. Only idiots would rig a count but not rig the exit polls to match.
The reason for the late-poll correction: they knew Kerry would win and so adjusted for a Gore-level win, they didn't expect a Kerry landslide.

[Febble]

15. The plot I think OTOH is referring to

is this one:



posted in this Daily Kos diary:

http://www.dailykos.com/story/2005/4/6/8028/83645

which also gives some p values.

The X axis is Kerry's state vote-count margin, so a swing state would be zero. I marked the "big three" (OH, FL and PA) in green. On the Y axis is WPE after conversion to "bias" (although this should really be done at precinct level - this was done on the state mean WPES from the E-M report).

Because of a quirk of the math, a negative WPE give a positive "bias" number so I have called that axis "Kerry over-estimate" - but before people jump down my throat, it means no more than that! That the sampling "over-estimated" Kerry's counted vote, which could just have easily meant that Kerry's vote was "under-estimated" by the count. But the axis label tells you the direction of the error. Zero is neutral, up is bad.

The other thing to bear in mind about this plot is that it represents the WPE alone. In other words the discrepancy between the proportion of voters for each candidate polled, and the proportion of fotes for each candidate counted. It doesn't include any error at the precinct-sampling level, which is why the plot may differ other plots you may have seen.

[Time for change]

32. I agree with your general skepticism

concerning the larger poll-election discrepancy in swing states as compared with non-swing states -- mainly for the reason you mention, i.e., that the difference is not overwhelming, as manifested by the p value you calculate of just under 10%.

But on the other hand (no pun intended), I do have some reasons for thinking that this is nevertheless an important issue:

1) Even a p value of 0.1 seems important to me when there is (in my opinion) so much a priori reason for suspicion.

2) I think that there are a number of reasons to believe that a one tailed chi square test would be more appropriate in this case.

3) The discrepancies in Colorado and Nevada were borderline statistically significant.

4) I believe that there is an important reason why election fraud would have involved non-swing states as well as swing states. Most importantly, I think that it was important to the people involved that Bush win the popular vote as well as the electoral vote. This would be much more likely if fraud was perpetrated in non-swing states as well as swing states. Thus, I think that the amount of fraud in a given state was determined by a combination of need (as manifested by swing state status) and opportunity (although I can't prove that, of course)

5) Regarding Febble's scatterplot that you mentioned, I don't think that that is the primary issue. That scatterplot looks at the association between WPE and Kerry vote margin. I think that it would be more important to look at the association between WPE and closeness of the election (or perceived closeness of the election prior to election day, i.e., swing state status) because closeness of the election is a better indication of need for fraud than is Kerry vote margin.

6) Lastly, the M-E report looks at WPE in swing states vs. non-swing states. This report shows a mean WPE of -7.9 in swing states, compared to -6.1 in non-swing states, and a median of -8.6 in swing states, compared to -5.1 in non-swing states. Of course, we don't know the statistical significance of those differences -- and we won't know unless and until we get some more specific data to look at.

[Febble]

41. Good point re perceived closeness

I have been thinking about that.

Do you have a suggestions for a data source? A poll of polls, perhaps?

What would be the relevant period, do you think?

[Time for change]

55. Perceived closeness

Well, first of all, in the months, weeks, and especially days before the 2004 election, I was so obsessed with following the polls that they became implanted on my brain. So I think that retrieving them from there would be one good way -- and I'm only half kidding. And when I do that I come up with something very similar if not identical to the 11 commonly mentioned swing states mentioned in my original post, item 7.

One could also look at the official results of the election. If you do that you get the same eleven states, with Colorado representing the biggest margin (5.7% for Bush), except that Oregon would have to be added to the mix (Incidentally, that may be significant because Oregon was one of the few states where Kerry actually did better in the official results than the exit polls , and Oregon I believe was done totally by mail, which as I understand it would have been very difficult to rig).

Or you could look at Mitofsky's exit poll margins. If you do that and you use the same 5.7% margin to determine what a swing state is, something very interesting happens (well, to me it's interesting): Some of the Kerry swing states disappear because they are tilted so far towards Kerry (PA, NH, MN), and some Bush states get thrown in (MO and NC).

Anyhow, I think that if you did a poll of polls you would find that the 11 commonly mentioned swing states would be the closest. I would look at all of the polls in the week before the election. There were a ton of them for most or all of those 11 states. In fact, to find out which states were perceived to be the closest you might not even have to look at the results of the polls -- just simply the number of polls that were done in the week before the election.

That assumes of course that if the electin were stolen, the strategy could have been revised appropriately in the days before the election. I believe that to be the case (don't ask me to prove that). I remember reading a thread not too long ago that implied that Karl Rove was very busy on election day and in contact with all sorts of people around the country... Well, I guess I'm beginning to talk like a conspiracy theorist now.

[Febble]

65. And of course

No exit polls for Oregon.

Can you send me a list of what you think the critical states were?

[Time for change]

67. I thought that exit polls were reported for Oregon

I have notes that say there was a 2.4 Kerry margin, but I can't remember where I got that from. Going back to look at the E-M report, I see that they didn't report anything there.

I list what I considered to be the swing states in item 7 of my original post, above. But as I said, there were a ton of state polls for these swing states in the week before the election. It must be possible to get them from somewhere, but I wouldn't know how to go about it.

[Time for change]

71. Oregon

I've been thinking some more about this, since it's a little embarassing to quote a poll and then find out that the poll doesn't exist.

Actually, I posted a report on this in November of 2004 (Actually my son, EOTE, posted it, since this was shortly before I joined the DU). The report was simply a state by state statistical analysis of the exit poll discrepancies. The data came from Jonathon Simon's article of November 11th, in which he noted that he was able to collect E/M poll data between 12:19 and 12:38 a.m. on Wednesday morning, November 3rd, because of a computer crash. The data included Oregon, which showed a 2.4% advantage for Kerry, which was 1.6% less than the official election results.

So, if Oregon was included at this time, why was it left out of the final M-E report? Since I'm a "conspiracy theorist", it really makes me wonder if Oregon was simply deleted because it ran counter to the rBr hypothesis. True, there were a few other states where Kerry did better in the exit polls than in the official vote count, but Oregon was a special case because of its all mail voting. Therefore, not only might it be used as an example against rBr, but also it could plausibly be deleted from the final report, perhaps, because of its anomalous voting procedures.

[Febble]

72. Well, my understanding

is that there were no exit polls in Oregon, because there was no-one exiting the polls - it was all postal voting.

But perhaps there were telephone polls?

Either way, Oregon would have been atypical, so I'm not sure you want to read too much into that. The whole point of the exit poll story is that exit polls are supposed to be accurate because they are done at the exits - in Oregon, whatever was done, it clearly wasn't an "exit poll" as such. That does not mean that the NEP may not have made a prediction for Oregon based on some other form of poll.

[Time for change]

73. But I would think that the main characteristic of an exit poll

is that only voters who actually voted are polled -- as opposed to pre-election polls, where formulas have to be devised in order to predict "likely voters". Couldn't a telephone poll have been done in Oregon to ensure that only persons who actually voted were polled?

I just went back to Simon's article. Oregon is there all right. 1064 responders:
http://www.scoop.co.nz/stories/HL0411/S00142.htm

But you're right that Oregon would have been atypical, so maybe we shouldn't read to much into it. But I do think it's worth thinking about.

[Peace Patriot]

46. Why weren't they?

"The exit polls aren't designed to audit elections. Maybe we wish they were, but they're not." --OnTheOtherHand

Why weren't they?

Fraud in 2000, so Congress sets out to "fix" it by flooding the states with money to purchase extremely insecure, unreliable, hackable new electronic voting doohickeys, owned and controlled by Bush buddies.

Numerous experts issue strong warnings about the insecurity and hackability of electronic voting.

Bush Cartel considered an illegitimate regime by many, due to 2000 election fraud.

The situation CRIED OUT for exit polls DESIGNED TO verify the election--as is ROUTINELY done in other democracies throughout the world, ESPECIALLY in situations of doubt about the powers-that-be who control the vote counting.

Just not in this one. Not even with major controversy here about stolen elections, and a new and controversial election system being tested out nationwide for the first time.

And WHY, deliberately, FAIL to give the American public the TRUE RESULT of the exit polls on election night?

Why FALSIFY the numbers to make it LOOK LIKE Bush won the EXIT POLLS?

They fooled ME, for about six hours. I thought Bush had won the exit polls. It was inconceivable to me (then) that all the major news organizations would FALSIFY the exit poll data in this critically important and hard-fought election (hard-fought at least by the Democrats), with so much controversy about Bush's previous stolen election and about electronic voting. But that's what they did. And that's how they earned my favorite epithet for them now: BUSH'S LAPDOG NEWS MONOPOLIES.

The fact that the exit polls showed a Kerry win all day long may be a point in favor of your assertion that Edison-Mitofsky was not out to defraud. But E/M may bear a more Byzantine and more obscure kind of guilt--i.e., the news monopolies DIDN'T COMMISSION them to verify the election, but rather deliberately commissioned them to conduct a kind of exit polling with as many fudge factors and future "talking point" excuses as possible, and E/M DIDN'T OBJECT (didn't say, "Wait a minute! This situation CRIES OUT for verification"--when they damn well knew it did). Thus, the news monopolies bear the major load of guilt for this journalistic crime, and E/M just went along for the money.

And, interestingly, that is just how E/M is behaving now--as if they bear the obscure, Byzantine guilt of having gone along. (--withholding data, proffering unsupported theories and excuses, hiding behind poll "design," leaking out bits and pieces of data to friendly parties, failing to even consider an hypothesis of election fraud). They are ACTING dishonest and guilty in some obscure way. That's what we're looking at.

To posit that they were "out to" defraud is a straw man, easily knocked down. But what about being a "go along" partner in a nefarious scheme NOT TO HAVE a verifiable election?

Considering what Bush's lapdog news monopolies did on the Iraq war (promulgating 100% baldfaced lies to the American people), and considering the tenor of their coverage of all things Bush, there can be little doubt that they wanted the Bush Cartel to remain in control of the country. And the only way that the Bush Cartel could remain in control of the country was by having an UNVERIFIABLE election--because, considering what-all Bush-Cheney had done (mass murder, massive theft, massive lying, massive federal debt, etc.), there is no way they could have won an HONEST election in 2004.

So, what do they do? They commission exit polls concerning which future commentators will be able to assert that "the exit poll aren't designed to audit elections."*

-----

*(E/M themselves have said, by the way, that there is no reason that their exit poll data cannot be used to verify the election. See Freeman's first paper.)

[Time for change]

52. I agree with your 1st major point but partially disagree with your 2nd

Whether exit polls were "designed to verify elections" or not, that doesn't mean that they can't be or that they shouldn't be. And in the situation that we now find ourselves in we certainly should use every means available to get to the bottom of this mystery.

Where I partially disagree with you is your apparent certainty that the abrupt change in the exit polls at about 1:00 a.m. on November 3rd is proof of an effort to defraud the American public. It is standard practice in surveys to "adjust" surveys to fit the actual results, when known, while preserving percentages that are not known. For example, Kerry won the Ohio exit poll with 53% of the female vote and 51% of the male vote -- a 4% victory. In order to "adjust" these percentages to reflect the official election results, while at the same time maintaining a similar male to female ratio, the pollsters changed the numbers to say that Kerry won 50% of the female vote and 47% of the male vote, a 2.5% loss to Bush. So a benign interpretation of this would be that they were just following a standard statistical practice.

But I only partially disagree with you on this issue because even though this is a standard statistical practice, it seems to me that in a matter such as this, with such huge public interest and public implications at stake, when the pollsters did this they should have explained to the American public what they were doing, along with a full release of the actual final exit poll results. I think that the fact that they didn't do this means that they consider this to be a private rather than a public matter, and in my opinion that exhibits a great degree of arrogance, if not an outright intention to mislead the American people.

[EFerrari]

14. Agree, have agreed, still agree. n/t

[Time for change]

31. Thank you for your insights Peace Patriot

I feel very similarly about the good majority of what you say here. I said in my original post that I wasn't going to say anything about non-exit poll related reasons why I believe that Kerry won the election. But you certainly cover a lot of them:

Voting machines with secret codes to count our votes;
Vigorous Republican efforts to prevent meaningful election reform;
The GA 02 Senate election, where pre-election polls differed hugely from the pre-election polls, and where the whole election was run on touch screens;
Massive maldistribution of voting machines and other violations of the civil rights of minority voters;
Massive new pre-election Democratic registration, especially in Ohio;
And Bush's consistent and continuing dismal poll numbers.
All of these things have added to my belief that Kerry won.

It is interesting that you mention that 86 out of 88 reported vote switching incidents favored Bush. I just recently did an analysis of "vote switching" reported to the national Electronic Incident Reporting System (EIRS), and my results were slightly different, as I found 87 out of 94 incidents to favor Bush: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=371211&mesg_id=371211 . Probably I used slightly different methods than the source that you quote. In any event, the conclusion is the same.

You also make some very good points regarding the swing states. Especially crucial is your point that whatever bias may have existed in the exit polls should have existed just as much in the swing states as in the other states. Therefore, a substantially larger discrepancy in the swing states casts much doubt on the validity of the rBr hypothesis. In fact, reading your comments about that made me feel that I should have emphasized this aspect more in my original post.

However, we also need to be careful as to how much importance to attach to this finding, because as OTOH points out, the difference between swing state discrepancy (between exit polls and official results) and the discrepancy in other states is not overwhelming. It is large enough that it contributes substantially to my believe that Kerry won the election. And I think that it is bigger than OTOH realizes (I will respond to his post shortly), but nevertheless it is small enough that this difference could have occurred by chance. Well, at least I think so. I also believe that this issue needs further investigation.

[Peace Patriot]

49. "Time for a change," I got the stat (86/88) from TIA. n/t

[snot]

8. THANK YOU!

from a non-statistician.

[Patsy Stone]

9. Excellent post.

Thanks, Time for change. It's good (well, not good, but you know what I mean) to see it all there in black and white. :thumbsup:

[autorank]

10. It certainly is TimeForChange. Excellent synthesis. Thank you!

This is very useful. I came to the same conclusion after reviewing TruthIsAll's outstanding and comprehensive work and sources outside of DU (including those you mention).

I was particularly impressed by by the commentary on 'febble:' "spending too much time on this may be harmful to our efforts to shed light on the main issues. " Understatement is one of your gifts. One wonders why the diversion has been presented here at this time. 'febble' recently announced that she has been/is now on the Mitofsky payroll (the very same person, it would seem, who confounded his own exit poll and with holds evidence).

It's a shame that so many fail to take a serious look, much like the Cardinals refusing to look through Galileo's telescope. The motivation of many may be similar to that of the Cardinals: to take a reasonable look at the various data sources and reason through the implications produces the very unsettling result of looking into the abyss as all of our cherished assumptions collapse into chaos.

We were robbed!


Excellent Summary of TIA/DU Proof of Election Fraud

[glitch]

11. Well done. You OP and Peace Patriots post #6 cover it for me,

especially Peace Patriots #1. You cannot divorce the bias of the election machine corporations from their results, since there are no actual realistic checks on their totals.

It's not statistics, but it is common sense to somehow factor "consider the source" into your analysis. This is what brings me to 99.9%. My gut instinct takes me the rest of the way.

Thank you for your post, nominated even though you don't need it at this point.

[Melissa G]

13. Excellent! Thanks for creating this comprehensive list! n/t

[Gelliebeans]

16. Thank you for taking the time

to put this vast amount of information in simpler terms.

To be honest, I have been so interested in this information but for lack of understanding on my part, I have stayed away from these threads. I know a great deal of work has been compiled on DU and other forums but it is very difficult to decipher and I have been hoping that someone would translate.

Thanks again Time for Change! I will keep this bookmarked.

[Febble]

18. Regarding 9:

I absolutely agree with you that my work is peripheral to the question. So peripheral, in fact, that there is no dispute between USCV and myself over the key issue in my work, which is that the WPE is a confounded measure of "bias", and it is confounded by the vote-count margin itself. And by "bias", I mean a measure of the extent to which the poll differed from the count - the "bias" could be in either the polling (by under-sampling Bush voters), or in the counting (by under-counting Kerry votes and/or over-counting Bush votes.

And that therefore re-analysis of the data was required if the true relationship between vote-count margin and "bias" was to be mathematically clear. All I did was to use a bit of algebra to remove the "vote-count margin" contaminant from the "WPE" and thus derive a "pure" measure of "bias".

But as you say, that is peripheral, and certainly not disputed by USCV.

This leaves us with two questions which are not peripheral:

1) What would fraud look like, in terms of the relationship between "bias" and vote-count margin?
2) Does the relationship between "bias" and vote-count margin (from Mitofsky's new plots) look like fraud?

Although the two questions are essentially similar, it is important to frame them both ways. There may be a number of answers to the first question (including "I don't know" - which has always been my answer), but answers to the second question might suggest answers to the first.

Oh, and I agree with OTOH that your post is an excellent summary!

[On the Road]

26. There's One Way in Which I Believe #9 May Turn Out to be Critical

USCV claimed that the exit polls were almost impossible to reconcile with the official votes in a certain category of partisan Republican precincts and in order to arrive at those results, there had to be overwhelmingly great Democratic participation in the exit polls (90% in scenario they discussed).

What the Liddle paper did was to refine the mathematical relationship and show that the participation rate should be closer to 70% than 90%. (That's still enough to warrant further investigation.) And while most of the data conforms to Liddle's curve, heavily Republican precincts are still an outlier -- the WPE actually increases in area where the model says it should decrease. That subset of precincts stands out and should be put under the microscope.

I'm open to correction if I'm misinterpreting the did (especially since the author is on this thread). What Liddle's paper did was to ensure that the discussion proceeded with a correct mathematical model.

What we've seen from Newsweek, Dan Rather, and countless other situations is that even if your conclusions are correct, a flaw in sources or methodology can be politically deadly. That's what peer review is for.

[Time for change]

38. Bias and vote count margin

First, thank you for the comment about my summary.

I have to say that I am confused about your two questions. I thought that my points 4 and 5 of my original post answer your first question. Together, I think that they suggest the unlikelihood of the rBr hyothesis, and therefore they suggest fraud.

But your questions seem to me to suggest that you believe otherwise. (nor am I am familiar with Mitofsky's new plots). If you could clarify that for me I might be able to respond.

[Extend a Hand]

21. Thanks from another non-statistician

[On the Road]

22. Thank You for This, TfC

and I say this as one of the official "doubters" here. I do believe that believers and doubters can and must work together.

Some of the controversy seems to be over how to argue a social science issue. What I see as a normal part of normal peer review and hypothesis testing often seems to be taken as a partisan attack.

A lot of the differences in perception seem to be related to your point #3:

3. Mitofsky’s explanation for the bias – the reluctant Bush responder (rBr) hypothesis

I think there's a common perception that Mitofsky looked at the election night results, had no idea how to square them with the official vote, and pulled something out of his ass to justify the official vote, namely the reluctant Bush responder hypothesis.

What I think of non-social scientists don't realize is that response bias is an absolutely core issue that infuses all polling and experimentation and always has to be controlled for. It may not always be in the headlines, but it's the forefront of every pollster's mind in interpreting odd results. Past examples from Mystery Pollster:

A fascinating pilot test looked for ways to boost response rates. The exit pollsters offered potential respondents a free pen as an incentive to complete the interview. The pen bore the logos of the major television networks. The pen-incentive boosted response rates, but it also increased within-precinct-error (creating a bias that favored the Democratic candidate), because as Frankovic put it, "Democrats took the free pens, Republicans didn't."

Studies between 1992 and 1996 showed that "partisanship of interviewers was related to absolute and signed WPE in presidential" elections, but not in off-year statewide elections. That means that in those years, interviews conducted by Democratic interviewers showed a higher rate of error favoring the Democratic candidate for president than Republican interviewers.

The specifics are not as important as the fact that the pollsters were not only aware of response bias, but were actively monitoring it and trying to ensure that it was minimized.

I'm also affected by my own major in experimental psychology. I spent senior year working on some projects related to perceptual defense, an subject which used to be hot in the 60s and 70s. In this case, it was easy to show the particular experimental effect. It was almost impossible to distinguish a real psychological effect from from response variables. There were several hundred academic papers published on the subject, most of which were attempts to remove response bias from the method (and most of which were unsuccessful). So it doesn't seem at all strange to me that there is so much discussion of response bias. And it doesn't seem at all strange to me that in a poll where only about half of the target was actually questioned, response bias could account for a 6% effect.

The key is to look at the pattern of data to see which hypotheses they support.

[Time for change]

36. Very interesting discussion

I think that this does give me a somewhat better perspective on Mitofsky's thinking in writing his report.

And I certainly agree with you that with a 53% overall response rate bias has to be seriously considered by any good researcher.

But I also strongly feel that a good researcher should consider other possibilities as well. Mitofsky appears not even to consider the possibility of election fraud, except to give it a seemingly off the cuff dismissal in his executive summary (See my post # 34 on this thread for my theory on that). I don't think that that is good science.

Furthermore, I also believe that point #s 4, 5, 6, 7, and 8 of my original post to this thread argue against Mitofsky's rBr hypothesis. Any one of them considered alone may not be very persuasive, but when considered together I think that they make a persuasive argument.

I feel that the bottom line is that Mitofsky, as evidenced by his report, appears not even to have considered alternative possibilities.

[On the Road]

43. Well, That May be True

Personally, Mitofsky may have given short shrift to the fraud arguments. I pay less attention to Mitofsky's attitude than whether the arguments from the data are compelling.

This discussion happens in public, but this is one reason why it's important for any arguments that are widely disseminated to be as bulletproof as possible. USCV is pursuing the right line of attack. They just needed a little refinement of their models.

If the most recent USCV paper had been the first thing to hit the streets, everyone would be looking at those heavily partisan Republican precincts that defy the trend. Instead, it's a third or fourth attempt to show fraud from the data, and each one carries less weight. In a more critical forum, their first paper might have been simply shot down rather than being improved.

[Time for change]

51. You lost me in your discussion of USCV

First you say that they are pursuing the right line of attack -- but then it seems to me that you're saying that they should have waited before publishing their first paper until it had reached the level of refinement of their last paper.

If that's what you're saying I disagree. I think that it was important to counter the Mitofsky report as soon as possible. And furthermore, I think that their first report was quite good. It didn't prove election fraud, but it certainly raised enough doubt about the election that is should have spurred much more investigation than we've had. And in a country with a reasonably independent news media I think that would have happened.

The point is that when it comes to a matter as important as this one, proof shouldn't be required before a vigorous investigation is undertaken -- any more than proof should be required to arrest someone for murder.

[On the Road]

58. I Wasn't Being Very Clear

There are different levels of publicity, and there's always a balancing act between speed and conclusiveness.

It would have been ideal if USCV could have published their original paper with a more refined methodlogy. That would have given it more credence within the polling community. But it was OK for it to be corrected at this stage -- the timing may have been worth it, as you point out. But it's better for wider audiences to become aware of the issue through their revised report.

What I hope USCV does now is break down those heavily Republican districts which deviate so much from the exit polls and which run counter to their model. It's possible that they're abberations. But paper precincts in partisan areas can easily inflate the official vote. To pursue that, more than USCV will be needed. Local activists and people involved in the recount efforts and legal challenges will be needed to do this. The stronger the statistical argument, the easier it is to get consensus and enthusiasm.

In the back of my mind, I'm a little miffed at USCV. I think they trumpteted their credentials too much (always a red flag). When you proclaim yor expertise so loudly, you place a special onus on yourself to be right the first time. I also thought USCV was condescending in their discussion of Febble, and tried to draw too much out of the statistical evidence. On the other hand, they zeroed in on this one set of anomalies with partisan GOP precincts which nobody else that I know of did. And it's the best argument from the exit polls around now.

[Peace Patriot]

48. Partisan attack?

"What I see as a normal part of normal peer review and hypothesis testing often seems to be taken as a partisan attack. "--ribofunk

Gee, I wonder why DUers and other defrauded and disenfranchised Kerry voters living in BushWorld would feel under partisan attack, with...

Bush partisans owning and controlling the source code that "counts" all our votes...

Bush voting machines changing Kerry votes to Bush votes...

Bush Baptists excommunicating church members who voted for Kerry...

Bush election officials shorting Democratic areas on precincts and voting machines...

Bush's brother purging 50,000 black voters from the voting rolls...

The Republican National Committee funding operatives to shred Democratic voter registration cards...

Bush "pod people" in Congress nuking the Democrats...

Bush packing the courts with Bush "pod people"...

The Bush Cartel stealing us blind...

Bush scientists denying global warming...

Bush freeper blowhards, and black ops subcontractors, running overt and covert Bush programs in every venue of power and news in the country, including DU...

...and Bush, Bush, Bush and more Bush all day long in the news monopoly radio & TV.

We would be idiots not to vet bloggers for hidden agendas, and can be excused for being a bit touchy about partisan attack, I think.

[On the Road]

50. Yup, That's the Pattern, All Right

That's precisely what has to avoided in pursuing the issue. If the goal is to pursue fraud, it is absolutely destructive to that end.

[On the Road]

25. Machine Type (#6) is Probably the Key Area of Controversy

Because the exit poll can never be repeated, the only way to argue for fraud from the data is to look for patterns in the data and show that they most closely fit your hypothesis, especially if it can shown on several different variables (eg, machine type, geography,
precinct size, partisanship). It's just as important to show that the data does not fit the null hypothesis, in this case that there was no fraud. (If both explanations fit the data, it does not constitute a very strong argument.)

The most convincing way to argue a situation like this is to have a hypothesis before the fact on what fraud would look like. I think there was an implicit belief that electronic machines were most vulnerable to fraud and that if there was fraud, it would show up in a higher WPE in precincts using Diebold or other electronic equipment. That didn't happen. Instead, there was an unexpected pattern:

- Paper ballot: -2.2
- Mechanical voting machine: -10.6
- Touch screen -7.1
- Punch cards -6.6
- Optical scan -6.1

Mechanical equipment, which is arguably the most difficult to hack, actually has the highest WPE. So when the exit polls are used as evidence of fraud, there are the following obstacles:

- The data did not follow the original scenario.
- Any explanation of the data must involve one of these three elements:
o The high WPEs in mechanical equipment precincts was due to fraud, implying greater fraud in what's traditionally viewed as a less hackable method.
o The high WPEs in mechanical precincts are due to some factor other than fraud. The follow-up question is why a factor which could account for a 10% difference in some places would be unable to account for a 6-7% difference in others.
o There was fraud across all kinds of voting equipment.

Based on the existing breakouts, none of these are particularly strong alternatives. The last case -- fraud in all areas across all machine types -- is sometimes argued here, but it lacks at least one critical element: a detailed scenario how how it could happen in the midst of so much Democratic and nonpartisan participation in vote gathering, reporting, and counting. It also has the weakenss of implying an effort involving a much larger number of people.

A detailed scenario is is not an unrealistic goal. Some of the early work on evoting had very detailed findings on Diebold operating systems and showed exactly how vote counts could be altered without leaving evidence.

Another example is that during the Florida recount in 2000, there was one punch-card county with something like 10,000 triple-punched ballots and very few double-punched ballots. The argument that this pattern could only be accounted for by fraud was supported by (1) basic probabilities, (2) an anaylsis of which candidates were punched, suggesting that many more Gore votes were spoiled than Bush votes, and (3) by showing a specific mechanical method by which large stacks of ballots could be quickly spoiled by punching an additional hole.

[TruthIsAll]

27. Here are some graphs which may help...

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x371891

[On the Road]

Thank You, TIA

I've seen that graph of yours but didn't think of it in this discussion.

Can you elaborate on what the data tells you? The correlation does not look especially strong (somewhere around 0.1), and errors are slightly higher in the middle than on the ends.

Among the states with large numbers of reported errors, there are the big three swing states (FL, OH, and PA) but also some strongly partisan states as well (DC, GA, NY).

[Febble]

29. Here's my plot

of state by within-precinct error:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=371726&mesg_id=371802&page=

on the post above

Again, it's the blue states that have the greatest error. I tried a quadratic function, and didn't get anything.

[TruthIsAll]

35. This graph should help even more..

[On the Road]

28. Dupe

[Time for change]

37. Unfortunately I can't argue this point very well

My impression was that the hand counting of paper ballots would be the most difficult to perpetrate election fraud on. If that is true, then the fact that paper ballots are associated with the least negative WPE would support the possibility of fraud. Of course we would need more detailed data in order to refute or confirm that with statistical tests.

You say that "mechanical equipment" would be the most difficult to hack. If that is true, then the table that I presented in item # 6 of my original post would, as you say, argue against the fraud hypothesis.

I don't know enough about voting equipment to argue this point. In fact, I don't know what is meant by "mechanical equipment" if it's not optical scan machines, touch screens, or punch card machines. I do know that in the case of those three types of voting machines, the votes are counted by machine. Isn't that the case with mechanical equpiment as well? And isn't the paper ballot category the only one where the votes are not counted by machines?

Anyhow, I certainly agree that those of us who strongly believe that this election was fraudulent and that Kerry actually won are nevertheless at a loss to explain precisely how it happened, other than that the machines that counted or registered the votes were programmed to make Bush win.

As long as these machines use secret software to register and count the votes, and as long as the public doesn't have the right to inspect these machines we'll never know for sure.

[On the Road]

47. I Took "Mechanical" Equipment

as referring to the huge old voting machines on which each candidate had a separate little level which you pushed down. When you drew the curtain back, the votes were counted and the levers were returned to normal. Totals were read off a kind of odometer in the back. Until 2002, that was the only kind I had used.

It's more difficult, but it's still possible to hack the machines. One way is to give one candidate a headstart on a particular odometer. Another is to jam an odometer. Another is for the person reading the totals to simply misrepresent the numbers. That's one reason nonpartisan or jointly monitored elections are so important.

I don't agree that paper is inherently less susceptible to fraud. The phrase "stuffing the ballot box" refers to a very common type of paper fraud. In 1962, Jimmy Carter personally tracked down and apprehended a stolen ballot box from a local Georgia election.

I'd actually like to see a study of cases of proven election fraud discussing who committed it and how it was done. I suspect that in most cases, only a minority was even aware of it. Like politics, fraud tends to be local. To me this suggests that the small partisan Republican districts uncovered by USCV are more likely to yield results than statewide deviations.

And BTW, congratulations on starting the most constructive and civil thread so far on election fraud.

[Time for change]

53. Fraud may tend to be local in general

But I think that in this case we are dealing with an administration that has the motive, the know how, and the lack of morality to take it well beyond the local. For example, we now have evidence that the Iraq war was planned well before 9-11-01. If they could do that they could rig a national election, I think.

Certainly you are correct that counting paper ballots by hand does not rule out the possibility of fraud. But I think it would be extremely difficult to perpetrate a national fraud if all votes were hand counted.

Also, there is widespread belief among people who understand our voting system much better than I do that with today's voting machines national election fraud is much easier than it ever has been. Here are just a couple of examples of what happened or may have happened on 11-2-04: http://www.rawstory.com/images/pdfs/CC_Affidavit_120604...
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=371211&mesg_id=371211
The fact that secret software programs are used to register and count our votes and that the Republicans are intent on keeping it that way further adds to the likelihood of a stolen election.

How mechanical voting machines fit into all this, I couldn't say. So you may have a point there.

And thank you very much for your last comment. Although I doubt that it is true, even if it is close that is saying an awful lot, given the thousands of threads that have appeared in this forum. And it is so gratifying to me to see so many positive comments about my post and this thread because I honestly believe that this is the most important issue in the world right now, so to think that I have made a worth while contribution to this effort is truly a humbling experience.

[On the Road]

57. The Vote-Flipping Reports Puzzle Me

not because they benefited Bush but because there are so few of them. Only 87 Kerry votes were reported flipped? To paraphrase some other posters: in order to believe that vote-flipping swung the election, you would have to believe that only one out of every 750 was reported.

If vote-flipping was sometimes done silently, why were there a few examples which were visible to the voter. This seems like a very serious red flag. Whatever interpretation you put on it, it seems to involve incompetence or misjudgment.

The material on Clint Curtis is chilling. However, if enough votes were switched to swing the election, I would expect it show up in the exit polls. As it is, mechanical devices had the biggest divergences as did Democratic areas like Cuyahoga County. It's entirely possible, of course, that there was an infrastructure in place to fraudulently swing a few thousand votes, but that it never had to be used because the early returns convinced Republicans that Bush won. It's possible that both sides are right.

I believe a lot of the differences on DU have to do with how to weigh evidence and argue fraud publicly. I frustrate a lot of people here by being a wet blanket, but my criticism is nothing to what fraud theories are exposed to in the general media. If an argument for fraud cannot withstand that kind of skeptical scrutiny here, it actually serves to discredit the entire fraud effort and should not be made public. At least, that's my opinion.

I do think that DU fraud discussions have changed in the last week or two. It's a good development. The most frustrating thing now is followup on many of the open items. Affidavits are one thing, but they are the first step. Sitting in Maryland, I honestly don't know how to pursue some of these things. I gave to the recount efforts several times, but after nothing turned up in New Hampshire and the legal cases were blocked in Ohio, everything seemed to fizzle. What does it take to get to the next step? Do we need to ask some folks from 2600 to find out what code is actually on those machines?

[Time for change]

61. Vote-flipping

True, in order to believe that vote flipping swung the election, the number of reported cases would have to represent only a small fraction of the total. But in the first place, the vote flipping could have been only one of several mechanisms to swing the election. And secondly, I believe that there are many reasons to believe that the reported cases are only a small fraction of the total:

Besides the fact that possibly only a fraction of the cases were visible, or that even if visible, perhaps most voters would not have observed the problem, I think that the biggest point is that very few voters would have reported it. As an epidemiologist I have spent several years dealing with infectious disease reporting systems, and it is routinely accepted that for many diseases the reporting rate is 1% or less. And this is in a situation where reporting is required by law. How many voters would even be aware that a reporting system existed? I am a partisan Democrat who donated to and did volunteer work for the Kerry campaign, and even I did not know about this reporting system until well after the election. And furthermore, if you look at the Appendix to the vote switching discussion you'll see that many of the individual reports noted something to the effect that this problem went on all day long.

If you think the material on Clint Curtis is chilling, how about this: I had the opportunity to meet him when I was invited to participate with a small group of people to lobby US Senators to officially object to the counting of the electoral votes on January 6th. He told me the story of the federal investigator who was working with him to investigate his claims (I can't remember his name, but I know it's documented). The investigator was very aggressive in his efforts, and then was found dead in a hotel room. It was ruled a suicide. I saw an article that purported to demonstrate that this was a murder rather than a suicide, but I didn't read it, so I can't tell you anything about it.

I believe that Mr. Curtis' testimony had to do with central tabulating machines, rather than individual voting machines. Again, I don't know much about the system used to count our votes, but if you're talking about central tabulating machines adding up the votes from individual machines or individual precincts, and if they're programmed to switch votes, I assume that it wouldn't matter if the individual machines were mechanical or touch screen or optical scan or punch card. This is really out of my territory, but it seems to me that Curtis' testimony had to mean something. Furthermore, there is no question that he put his life in danger by testifying before the House Judiciary Committee. How many other computer programmers found themselves in similar circumstances, and how many of them would have testified to this at the risk of their lives? In other words, I think that the vote switching did show up in the exit polls, especially the Ohio exit polls, which had Kerry winning by 4.2%.

I appreciate your criticisms. I see them as constructive, and therefore helpful to our cause, for the reasons you state.

What is the next step? I wish to hell I knew. I'm just trying to be helpful by trying to understand better what's going on, contributing to the discussion, and doing some volunteer work for the Ohio Project and Voters Unite. I'm also trying to come up with some great insight into this whole thing, but it's really an uphill battle.

[On the Road]

62. Your Perspective in Epidemiology is Interesting

Maybe it's completely reasonable that almost all of them would go unreported. I wouldn't have thought so. And I had forgotten about the reports that it had been going on all day. That could mean a difference large enough to cause a noticeable difference in results. Do you know which precincts have had the problems reported? Any in Ohio?

As far as next steps go:

(1) In the case of the vote-flipping reports, the next step would seem to be to compare the official vote in those precincts with the vote in previous elections, and to map those changes against adjacent or similar precincts. That information should be available somewhere. The goal would be to get a scatterplot which looked predictable except for the vote-flipping precincts, which would stand out like a sore thumb.

(2) I also didn't remember the details of the Clinton Curtis situation. Changes at the central tabulating stage are the one place which could cause the kind of systematic deviation that was seen in Ohio. I don't have difficulty in believing that this could be done undetected in Diebold precincts. I do have difficulty in seeing how this could be done statewide with optical and lever devices, since AFAIK they have to be counted at the local level.

But maybe there was a way to skew the whole state without detection. I don't know. In this case, the next step would seem to be to create a model of how this could be done without detection. Not so much on the computer side, but on the human side. Is it reasonable to think that in dozens or hundreds of reporting locations, the officially announced vote diverged from the vote counted at the local level? Maybe, like unreported vote-flipping, this is more likely than it looks. But a scenario or model has to be created based on existing procedures? Know anyone in Ohio who works in statewide elections?

(3) The last avenue would be to start with the exit polling data. It's in some ways the most difficult, because the results are all over the place and the differences with the official vote affect so many precincts. (In fact, to me the biggest argument against fraud is that the divergences are so uniform geographically.)

However, there are a lot of big outliers in both directions. They would include the group targeted by USCV but also a lot of others. Some of those divergences may be due to fraud. If those outliers can be identified, then voting history can tell us whether the official results are more consistent with the exit polls or with past voting history. But only Mitofsky can provide that information. Maybe there's a way to get him to provide the names of the outliers. I don't know.


I don't expect a purely statistical argument based on the exit polls to go anywhere. I still believe that local fraud is more likely than universal fraud. But the statistics could suggest where to look for fraud on a local level. If organized fraud can be shown in any individual location, that will open the door. That's what I'm personally hoping for.

[Time for change]

70. Next steps

I don't know about the precincts, but I can tell you that 42 of the 87 reports were from the three strongly Democratic counties in southeast Florida (Palm Beach, Broward, Miami Dade), and 8 were from Mahoning County, OH. Mahoning County is the only Ohio county listed by the database I used from Voters Unite! as touch screen. There are also about 6 other Ohio counties that are listed as "electronic - other", though I don't know what that means.

With regard to your "next steps":

(1) This may or may not be practical, since the vote flipping incidents probably represent such a small portion of the total. In any event, I don't have access to precinct data. Do you know how that could be obtained?

(2) I certainly agree, but I wouldn't know where to start with this one, since I don't know enough about voting machines. People with expertise in this area need to get involved with this one.

(3) Again I agree. I think that USCV is working on this one.

[eomer]

77. Some comments

"Not so much on the computer side, but on the human side. Is it reasonable to think that in dozens or hundreds of reporting locations, the officially announced vote diverged from the vote counted at the local level?"

The theory that has been thrown around is that the tabulation machines at the county level were rigged. There are 88 counties in Ohio. But it would not have been necessary for someone in each of the 88 counties to be involved because the personnel at the county BOEs have no idea about the technical side of the tabulation equipment. They are totally dependent on the vendors. So a small number of people at the vendor (possibly as few as one) could have rigged the counting software.

During the recount there was suspicious activity by vendor tech representatives who made visits to the BOEs. In some cases the tech seemed to use false pretenses to justify working on the tabulation equipment. In other cases the tech coached BOE workers on how to fudge the count. In still other cases there is evidence that tabulation equipment was accessed and modified by way of remote connectivity.

All this has been discussed at length so I assume you are relatively new to this forum. If so, welcome! Or maybe you've visited less frequently. Either way, there has been a lot of discussion and you may want to do some searching to find old threads on these subjects.

While I'm at it, you mention the Clint Curtis situation. The gist of it is that Curtis created a prototype, not an actual working hack. He apparently created a mock-up of a touchscreen voting application using Visual Basic that would run on a regular PC. Then he wrote a hack of his own voting application mock-up. The code he wrote would not have been usable "as-is" to rig any actual voting system. It was only useful in that it illustrated the concept. So the information from Curtis doesn't give any clue about which type of voting system - touchscreen DREs, optical scan tabulators, central tabulators, etc. - may or may not have been hacked. It just shows general intent to hack voting software on the part of Tom Feeney.

[Peace Patriot]

76. The vote-flipping was largely invisible to the voter...

"The Vote-Flipping Reports Puzzle Me / not because they benefited Bush but because there are so few of them. Only 87 Kerry votes were reported flipped? To paraphrase some other posters: in order to believe that vote-flipping swung the election, you would have to believe that only one out of every 750 was reported.

"If vote-flipping was sometimes done silently, why were there a few examples which were visible to the voter. This seems like a very serious red flag. Whatever interpretation you put on it, it seems to involve incompetence or misjudgment." --ribofunk

------

On touchscreens, the voter starts with the presidential election, votes, then "turns the page" (electronically) to the next set of candidates--the lower ticket candidates--then the next, then non-partisan races, then initiatives, etc., each time "turning the page" electronically. The voters who reported Kerry votes being changed to Bush votes were very meticulous voters, who "paged" back and double-checked their votes. In several reported incidents, they "paged" back to the beginning (electronically), found their vote for Kerry changed to Bush, restored their Kerry vote (or thought they did), then checked other votes ("paged" forward), then went back and checked their Kerry vote again, and AGAIN it was changed to Bush.

Some who did this commented on how much trouble it was, and that other voters probably didn't do what they did. Hypotheses: Most voters didn't check their vote for president before leaving the polling booth.

Other kinds of ballots make double-checking easier. In most, you have it right there in front of you. Your vote is a physical object--a ballot--that you punch, or that you fill out and put through the optiscan device. The whole thing is in front of you. With touchscreens, that is not the case. When you vote on one part of the ballot, it then vanishes. You have to go to some trouble to check back.

Also, it's likely that many voters who checked back through the "pages" and NOTICED that their Kerry vote said Bush, would have thought it was THEIR mistake, changed it to Kerry, and then forgot about it--probably didn't check again, and probably didn't think it was anything untoward.

Because of poor info on touchscreens and on who owns and controls the programming code, most voters who even noticed the changed vote--who bothered to "page" backwards--would not likely have thought of election fraud, would likely have blamed themselves not the machine, and would not likely have reported it, especially if they didn't check it twice. Also, a) they may have been in line for hours--pressure to move along; b) they may have been in a hurry; c) there may not have been help available--or hostile polling officials; d) they may not have known who to report it to; e) incident reporting may not have been very well organized in their area--lack of public education, over-stretched poll monitors.

90 reports (or whatever it was--close to 100) is A LOT. It points to programming as the method of fraud. ALSO, the huge preponderance of the incidents being against Kerry is a red flag. If this was simple "machine failure" or "human error," it would be more random.

[Time for change]

78. Thank you for elucidating this Peace Patriot

I tried, but I don't know that much about the voting machines. I just thought that, as you say, the ratio is very suspicious and probably represents just the tip of the iceberg, and that's why I thought it was worth reporting on.

Exposing what went on in the 2004 election is going to require the efforts of many, with many different kinds of skills. I can discuss most of the statistical anomalies, but I can't begin to explain how the votes were shifted.

For example, Ribofunk made a point about mechanical voting machines in post # 47 above that I couldn't answer very well -- I could only speculate. Why the large negative WPE in precincts that use mechanical voting machines? Wouldn't they be harder to hack than electronic touch screen machines? I speculated that it might have to do with the central tabulators (as Clint Curtis discussed in his testimony before the House Judiciary Committee). Is that a good explanation? I don't know.

[yowzayowzayowza]

40. Thx for the digest...

Itz easy to get lost in the acronyms! :kick:

[RonB]

56. Sunmmary of the USCV/E-M and Liddle Debate

Dear Democratic Underground Readers,

I have asked to post on Democratic Underground. I hope this "summary of the debate" is useful.

The following summary was written as a response to a post complaining that USCV (www.uscountvotes.org) was making a fuss about minor discrepancies, implying that the E-M case had been proven by the scatter plot of the WPE-index (Ln(alpha) – the “Liddle” index) shown by Mitofsky at AAPOR.

The response is, almost in full, as follows:

a) USCV initiated the study of partisan exit polling response rates in order to show that implausible partisan response rates are necessary to generate the aggregate tabulations released in the E-M report (see USCV March 31 updated April 12 report). The report develops a methodology to calculate Kerry and Bush voter exit poll response rates (K and B) which when multiplied by their respective reported vote shares (k and b) and added up (assuming negligible independent vote) give an overall exit poll response rate (R = Kk + Bb). Table’s 2 – 4, p. 11 and p. 25 of this report, show that implausible variations in K and B from “representative” (mean valued) precincts would be necessary to generate the mean WPE’s (E) and overall response rates (R) shown in the E-M report.
b) Elizabeth Liddle, who was an active participant in the USCV discussion list at the time discovered that when simulating the logged ratio of possible K/B values – based on the derivations developed in the April 12 USCV report - (Log (K/B) - the log was added – I believe at the suggestion of another participant on the list - to make the ratio symmetric), over partisan precincts, a noticeable “inverted U” asymmetric shaped WPE pattern emerged. This was particularly noticeable when simulating an (implausibly) high 2:1 K/B ratio (see Liddle figure 1 - displayed by Warren Mitofsky as part of his AAPOR presentation).
c) Liddle may have unknowingly reinvented the “wheel” algebraically but her index (the WPE_Index displayed by Mitofsky at his AAPOR presentation) is equal to Log(K/B) where all these terms are based on the derivations developed in the April 12 USCV report (see Appendix C –D in latest, May 12, updated May 21, USCV report). The importance of this is not only one of correct attribution of credit (appreciation of the value of an analysis should not be dependent on the conclusions that are derived from it!), but also to point out that USCV and Liddle have been looking at the same variables (partisan exit poll response rates) derived in a mathematically equivalent fashion.
d) E-M seems to be mistakenly, or deliberately, trying to create the impression that the Liddle analysis is based on some new, hitherto undiscovered “artifact” or “confounding”, that resolves the debate and shows that the mean rbr hypothesis can, after all, in spite of appearances to the contrary, explain the data. This, unfortunately, may partly be a result of the appearance of Liddle’s paper, in which she derives “alpha” (in a rather complicated and convoluted manner) apparently without reference to the equations for K and B presented in earlier USCV reports. (The obvious way to get alpha, once K and B have been derived, is in one or two easy steps as is shown in Appendix C of May 21 USCV report.) In addition, though she refers to the USCV paper in at least one footnote, Liddle unfortunately does not point out that the USCV analysis was based on these very same K and B partisan response rates patterns that she investigates in her paper.
e) In any case, based on her finding of the “asymmetric inverted U” pattern, Liddle came to an opposite conclusion to that of USCV. She surmised that this pattern indicated that the E-M data could have been generated by an unvarying mean partisan response “bias” which she defined as K/B.
f) How could she have come to such a different conclusion? The USCV reports used E-M reported mean WPE and R values to calculate K and B levels in different categories of precincts.. These K and B values diverged markedly in implausible ways across partisan precincts. Liddle, on the other hand, simulated (K/B) to get the suggestive pattern displayed by Mitofsky above. By looking at an exaggerated (by assuming a 2:1 K/B ratio equal to an alpha of 2) simulation of the K/B ratio she produced an “inverted u” shaped WPE graph that seemed to produce an “inverted u” WPE graph that seemed curved enough, and asymmetric enough, so that it appeared that it could approximate the know E-M reported WPE outcomes.
g) In response to Liddle’s pointing out this asymmetric inverted “u” WPE pattern, USCV added Appendix B (which appears in both reports) which derives this pattern from “Differential Partisan Response” (w = B – K). This appendix shows why the “inverted u” pattern appears, why WPE will be at a maximum perfectly competitive districts (k=b=.5), and why differential partisan response (w) will be equal to WPE (E), if k=b=.5 and R = .5. Appendix B also points out that mean calculations that were already done in the USCV (April 12) report show that the “inverted u” finding makes the jump in WPE to -10% in high Bush precincts even more implausible (as highly partisan precincts should have smaller WPE), and that to get the WPE’s for the other categories of precincts, the calculations in the report show that w would have to swing from 40% to an absolute minimum of 20.5%.
h) In addition, Appendix B shows that the “asymmetry” of the “inverted u” WPE curve – which gives a larger WPE in high Republican precincts (see Liddle Table 1 and Mitofsky presentation) that seems to be consistent the (much higher) WPE of high Bush precincts in the E-M data, is a mathematical result of linking an absolute difference (WPE) measure to a ratio measure (alpha). This “mathematical nit” cannot possibly explain the dramatic asymmetry in the E-M data (see the WPE’s generated by a constant alpha = 1.15 in Table 2, p. 19, May 21, USCV report). Moreover, if an absolute difference “differential partisan response” measure is used (w= B-K), even this small asymmetry disappears altogether. Only with highly magnified levels of Alpha (such as a 2: 1 ratio representing alpha=2) will this small effect look significant.
i) This debate with Liddle went on for some weeks. USCV members did further calculations based on means and medians with “alpha” (=K/B) and showed that using either means or medians “alpha” would have to range from below or almost equal to 1 in high Kerry precincts to 55-58% above in high Bush precincts to get the E-M reported WPE outcomes for representative precincts (see Table 1 in Appendix F, p. 19 of most recent May 21 USCV report.)
j) USCV showed that the overall response rate (R) levels that would be generated by representative precincts under the E-M “alpha” hypothesis of K =0.56 and B=.5 were mathematically infeasible in high Bush precincts and highly implausible in high Kerry precincts (R would have to drop to 16.9% - 29.2% in these precincts). Note that Liddle’s index is simply a ratio index that says nothing about overall response rates. Though K and B can be generated from r (=(B+K)/2) and w (see Appendix B) , they cannot be calculated from alpha (=K/B) alone. Another parameter like r is needed to get K and B and overall response rates (see derived equation in Table 4 of Appendix 4.
k) Liddle attempts to address this in her paper by noting that the E-M report states that the overall response rate (R) changes are not significant, implying that the bias (alpha = K/B) is all that needs to be looked at. But this skirts the issue. If a unvarying mean bias (mean constant alpha) hypothesis can only be sustained with radically divergent and implausible overall response rates, the hypothesis cannot explain the data, which as noted, show small (and possibly statistically insignificant) variations in response rates. USCV’s first report also showed that these small response rate variations contradict the E-M hypothesis.
l) None the less, Liddle was not swayed that a constant mean response bias (alpha) could not possibly (with any reasonable degree of chance) explain the E-M data. A key issue at this point was whether “aggregate” calculations using means and medians (such as the ones the USCV had used in its first report and in Appendix B) could adequately represent outcomes from randomly varying precinct level simulations. Recall that Liddle’s analysis at this point was based entirely on hypothetical simulations. In response USCV did some output simulations – trying to match E-M WPE and response outcomes with constant alpha. These also showed that matching E-M reported mean and median WPE levels, and over-all response rates with constant alpha was highly improbable to impossible (see Appendices G and H in the May 21 USCV paper).
m) Liddle remained unconvinced. She went ahead and published her paper, which she had already written about on several web sites. Her paper was hailed as holding the key to saving the constant mean bias hypothesis. The fact that it was based on the very same data (and pretty much the same analysis that USCV had be doing to show the opposite) was overlooked.
n) USCV felt pressured to respond to the Liddle paper by releasing the May 21 paper, and sending representatives to the AAPOR conference. This new USCV paper in addition to the calculations in k) above, also included an input simulation that showed that the E-M hypothesis could not explain the E-M outcomes, even under the most extremely (favorable to the hypothesis) precinct distributions (see Appendix I ). Moreover, these simulators (one of which is in completely transparent Excel form) have been put on the USCV website so that anyone can verify these results.
o) The Liddle and E-M hypothesis had been rejected at this point in at least three different ways.
a) Analysis of “representative” mean and median precincts showed that it was mathematically infeasible or highly implausible (Appendix F, May 21 USCV report).
i. Alpha’s necessary to produce the E-M data change by more than 62% (from mean calculation), or 52% (from median calculation – see Table 1, p. 18, May 21 USCV report).
ii. An alpha of 1.15 (representing an even greater “bias” than the 1.12 hypothesized by E-M and Liddle) is unable to generate the large E-M reported WPE values for high Bush and competitive precincts, and the small WPE for high Kerry precincts (from mean calculations – see Table 2, p. 18).
iii. Under the E-M hypothesis of K=.56 and B=.5 (so that alpha=.56/.5=1.12), the overall response rates (R) for high Bush precincts would have to be a mathematically infeasible -564.4% (from mean calculations) or also infeasible 423.3% (from median calculations). Overall response rates in high Kerry precincts would have to be a highly implausible 16.9% (from means) or 29.2% (from medians), see Table 4, p. 19.
b) “Output simulation” shows that with a high degree of certainty, matching E-M mean and median response rates, and overall response rates, requires significant unexplained changes in K, B, and alpha (Appendices G-H, May 21 USCV report, especially table on p. 20).
i. non-uniform mean alphas that change by at least 31% across partisan precincts.
ii. non-uniform partisan exit poll participation rates (K changes by at least 16%).
iii. High Bush precinct Kerry voter exit poll participation rates that are much higher than Bush voter participation rates (by at least 16% and sometimes up to 40% to 60% higher – p. 21).
c) “Input simulation” shows that it is impossible, under the most extreme favorable to the Hypothesis circumstances, to get E-M reported results from an E-M constant mean K=.56 and mean B=.5 hypothesis (Appendix I, May 21 USCV report). In particular this simulation shows that the E-M reported:
i. High Bush precinct mean WPE
ii. High Bush precinct median WPE
iii. Low Kerry precinct mean WPE
iv. High overall response rates in high (b>.8) and moderately high Bush precincts.
Are all unobtainable under the E-M hypothesis.

p) None the less, E-M, based on the Mitofsky AAPOR presentation and your comments, has embraced the Liddle analysis as providing conclusive evidence for the constant mean bias hypothesis. The key part of this argument appears to be the WPE by LN(alpha) linear correlation analysis presented by Mitofsky at AAPOR that we have been discussing. The question of how such a simple correlation analysis could trump the data already presented by USCV regarding representative precinct mean and median partisan and overall response rates across partisan precinct categories, and the more recent conclusive simulation outcomes, seems to have been lost or deliberately ignored.
q) It should be noted that the E-M report itself simply supplied tabulations and discussions of “factors” that could influence exit poll response and bias and an assertion of the E-M, that partisan response rate of K=.56 and B=.5 could explain all of the WPE error (p. 31 of E-M report). No solid statistical evidence (for example a multiple regression analysis actually showing that the factors can explain the WPE patterns, and that these factors result in .56 and .5 response rates, is offered to support this hypothesis. By the way, I agree with you that the term “reluctant Bush responder” is inaccurate. I was simply using the term as it has been coined by the media. One of the ways to get beyond the simplistic psychological metaphors, to the real factors that influence response rates, is to do the serious multi-factor analysis.
r) Mitofsky claimed, when I queried him about this at the AAPOR conference, to have done the regressions but not released them. This is doubly unacceptable! If they were done, the public has a right to see them! However, I am somewhat skeptical that they have been done, or at least done in a thorough and complete manner, as cursory analysis shows that K=.56 and B=.5 (generating an alpha of 1.12) could not possibly explain the relative magnitudes of WPE shown in the E-M report. See Table 2, p. 19, May 21 USCV report: alpha had to be increased to at least 1.15 to get WPE’s in range of the E-M data. It seems to me that this hypothesis, stated with such certainty on p. 31 of the E-M report, could not have come from an in-depth and serious statistical analysis by some of the “best analysts” in the country!
s) As I stated in my earlier post, the kind of tabulation, and now linear correlation analysis, that E-M has released to the public, would never pass muster as supporting evidence in any kind of serious academic journal (including one that I am an editor of). The gullibility of the media in support of the constant mean bias hypothesis without any serious evidence for it has been a travesty. The notion that “these things take time” etc. is also unacceptable. The credibility of our election system is an extremely important national issue – it should not take six months or more to provide a serious analysis (especially if some of the “nations best” analysts have been looking at it) of such an important issue. Moreover, there is no reason that private business contracts or personal confidentiality should trump critical public interest in this data. There are ways to release this data that protect confidentiality (as has already been done for Ohio). There is no sufficiently important or legitimate reason for E-M not to release the data and very good reasons, relating to a minimal sense of public responsibility and survey ethics, for E-M to immediately release the data without further delay. This is what I meant by “E-M needs to release the data”.
t) It may seem a bit beside the point, after all this to be debating whether a single linear correlation analysis of the E-M data is consistent with the non-varying mean exit poll response bias hypothesis. After all, the statistics (mean, median, and absolute value WPEs, and mean overall response rates) for the different precinct categories have been calculated and reported on. These are clearly highly divergent. Moreover, the influence of precinct partisanship has been eliminated from these data by calculating the direct K and B response rates (all done in the April 12, USCV report) and these show that implausible changes in K-B and K/B (the log of which is the Liddle/Mitofsky WPE-index) are necessary to generate these data from “representative” precincts. Why then should we be debating whether an insignificant linear correlation between this WPE-index and precinct partisanship shows that it unvarying?! We have already done the calculations and the analysis showing that this is not the case! For those concerned about “aggregation bias” in using “representative” precincts, we have shown that this hypothesis is highly implausible, if not mathematically infeasible, with precinct level simulations as well – see point o) above.
u) I submit that whether or not the scatter plot, considered as whole, produces a significant linear correlation or not, is under these circumstances, irrelevant. After all, a zero correlation can be produced with any number of non-linear variations. In this case the range of alpha’s (taking natural logs of mean and median alpha columns in Table 1, p. 19, May 21, USCV report) is (going from low Bush to high Bush quintiles), from means: - 0.0166, 0.1448, 0.1704, 0.1414, 0.4626, and from medians: 0.019, 0.137, 0.168, 0.141, 0.438. This would seem to imply a lot of variation and a positive correlation (with b). However, evidently, because of the very small sample sizes for the highly partisan precincts, the “inverted u” (not flat linear!) shape of the alpha from 90% of the data that is clustered in the less partisan precincts is sufficient to generate a flat zero correlation.
v) This “inverted u” alpha (not evident when drawing a straight line through the scatter plot) suggests that constant alpha is insufficient to generate the large WPE for competitive precincts (-8.5%) relative to the lower WPE levels (-5.9% and -6.1%) for the less partisan districts. This in it self may rule out unvarying alpha. This will depend on the significance levels of these differences – but given the very large sample sizes for these precinct categories, small differences in mean alpha levels (of +0.03 or so) are likely to be significant.
w) You make claim that the only unusual thing about the scatter plot data are four high Bush outliers in the high Bush quintile that are not offset by any high Kerry outliers, and what’s the big deal about four points, though you support investigating these precincts. First, it is curious to have only 40 high Bush precincts compared to 90 high Kerry precincts (measured by reported election outcome) in an election that Bush won by 2.7%? One would expect rough equality or a slightly larger number of high Bush precincts in a representative sample. This looks strange and may indicate that some other high Bush “outliers” have already been dropped from the sample. Second, even if for some reason there were less than half the number of high Bush precincts in the sample, four outliers represents 10% of a sample of 40. If 10% of all of the high Bush precincts in the country were corrupted, this could represent a very serious problem.
x) Moreover, all of the other ways in which these data are not consistent with constant mean alpha cannot be addressed by simply removing the four outliers. Whether or not the high Bush outliers are removed, the USCV reports have shown, for example, that the E-M hypothesis is not consistent with the high Bush median (which presumably would not be greatly affected by removing outliers) with the high Kerry mean, and with the high, and relatively high, Kerry overall response rates.
y) However, I reiterate, the basic point about the correlation analysis of the scatter plot is that it is wholly inadequate. Such an analysis “of the whole” will not provide detailed (or accurate if the variation in alpha is non-linear) information about what’s going on in the most interesting extreme partisan precincts where the constant mean bias hypothesis is really put to the test. As has been shown above, detailed analysis of the E-M constant mean response bias hypothesis breaks down in multiple ways particularly in these kinds of precincts.
z) The constant mean bias conjecture remains an unsupported (and largely inconsistent with the data that has been made public) hypothesis. Six months after the election, we still have no serious explanation for the large exit poll discrepancy. The shoddy (if I could borrow a term) and inadequate analysis (claiming for example that tabulations and linear correlation analysis are sufficient to support the E-M hypothesis) that has been released to the public has just deepened the uncertainty about what happened in the 2004 elections. I don’t see how this could be viewed as anything other than a national disgrace. The volunteer work of USCV, and other citizen activists who are deeply concerned about the credibility and/or integrity of our electoral system, and have refused to be satisfied with this pabulum, may, in fact, be the one glimmer of hope in this mess.

I hope this answers some of your questions and that you can convey, at least some, of my, and my colleagues, frustration and outrage over this situation, to people who have the power to do something about this.

Best,

Ron Baiman

[eomer]

59. I started a new thread for Ron's post,

so please post any discussion there rather than replying here.

Here is the new thread:
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x372464

[TruthIsAll]

66. TFC, welcome aboard. Your posts are elegant examples of how rationalism

Your posts are a breath of fresh air. They are elegant examples of why rational, logical analysis will always win the debate.

You see right through the obfuscations while summarizing your findings perfectly. It's too bad you were not around DU these past six months. We have seen many naysayers come and go. They would have left more quickly if you were here to lend a hand.

As it is, we're down to the final stretch. The few still afflicted with cognitive dissonance will soon be left to contemplate the absurdity of their arguments. Ultimately, most of them overcome the dissonance and succumb to reason. They see the light of truth, thene vanish from DU. Others never do. They pop up from time to time with personal attacks. That's all they know.

The evidence of fraud is now overwhelming. We just need some more truthseekers like Lampley and Koehler. We know they're lurking here. Postings such as yours help get their motor running.

[Time for change]

68. Thank you very much TIA

Actually I did become a member of DU shortly after the 2004 election. I did post frequently between then and January 6th, but after that my posts became much less frequent. After the inauguration, most of the air seemed to be let out of the balloon.

Thank you for your continuing efforts and wealth of statistical information. If it wasn't for the DU I don't know what I'd do, because it sickens me to listen to the MSM.

[deacon2]

75. Exemplary post and discussion thread - my thanks to everyone!

I found the intellectual sparring and development of the debate compelling from all sides, but perhaps a little too polite. I kept waiting for someone to be called a no good son of a bitch. Guess it's the Irish in me...

[Time for change]

79. Thank you very much deacon2

There was a fair amount of disagreement on this thread, specifically some "doubters", meaning people who don't share the strong belief that some of us have that Kerry won the 2004 election. I found their input to be very constructive, helpful, and conducive to good discussion.

This is a very emotional issue for many of us, and there is a natural tendency to get upset when we hear arguments that don't coincide with our beliefs, and sometimes we therefore resort to personal attacks.
I think it's important to resist that temptation.

I feel confident that all of the persons who posted on this thread are Democrats, or people with progressive political values. We are up against a very powerful and, I believe, evil enemy. If we attack each other when we disagree with each other I think that weakens us in our efforts to fight the real enemy.